From elli.chatzopoulou at incf.org Tue Jul 1 11:38:03 2008 From: elli.chatzopoulou at incf.org (INCF - Elli Chatzopoulou) Date: Tue Jul 1 12:46:13 2008 Subject: [Comp-neuro] Post Doc Position for Development of PET Imaging Reconstruction Software Message-ID: <4869FAFB.3020304@incf.org> *Stockholm Brain Institute* offers state-of-the-art facilities for PET imaging with the High Resolution Research Tomograph (HRRT) scanner and state-of-the-art analysis possibilities with its IBM Blue Gene supercomputer located at PDC, KTH. This two year full-time position will develop advanced analysis tools to enhance the accuracy of quantification and to optimize the temporal resolution of brain PET studies. The implementation of such tools will be achieved through the use of parallel programming on the Blue Gene in collaboration with IBM. *Qualification* Candidates should have a background in computer science or a closely related field and have experience with high-performance parallel computing, PET reconstruction techniques, and image analytics. Candidates interested in disseminating high-performance computing techniques to other users within SBI will be preferred. *Employment* Form of employment: Two year temporary position. Start date: According to agreement KTH aims to employ a diversity of talent and thus welcomes applicants who will add to the variety of the University, especially as concerns its gender structure. *About us*: KTH is the largest technical university in Sweden. At KTH education and research cover a broad spectrum within natural sciences and engineering, as well as architecture, industrial engineering and management, urban planning, work science and environmental engineering. There are circa 12,000 full-year undergraduate students, 1,400 postgraduate students and 3,100 employees. CSC is one of Sweden?s most advanced and successful research and education institutions in Information Technology. We work with education and research in Numerical Analysis, Computer Science, Media Technology, Human-Computer Interaction, Speech Technology, Music Acoustics and Languages at KTH and at Stockholm University (SU). For more information see http://www.csc.kth.se PDC is a major Swedish high-performance computing center located at CSC. PDC is also one of the members of Stockholm Brain Institute (SBI) a new brain research institute formed jointly between Karolinska institutet, KTH, and Stockholm University in the area of cognitive and computational neuroscience. For more information please note our web-pages: about KTH: www.kth.se about PDC: www.pdc.kth.se and about SBI: www.stockholmbrain.se *Application* The application -including CV and two letters of recommendations- should be sent via ordinary post to: KTH, CSC Att: Susanne Bergman 100 44 Stockholm, Sweden Applications via email to: susanneb@csc.kth.se Deadline for applications: 2008-08-25 Quote the following reference number: D-2008-0267. *More information* Please contact: Gert Svensson, PDC; KTH Phone: +46 8 790 78 84 * gert@pdc.kth.se or Andrea Varrone, KI, Phone: +46 8 517 750 43. * andrea.varrone@ki.se From jiang at gradschool.uni-luebeck.de Tue Jul 1 17:04:11 2008 From: jiang at gradschool.uni-luebeck.de (Chaoqun Jiang) Date: Tue Jul 1 17:11:25 2008 Subject: [Comp-neuro] Call for applications: PhD positions at Graduate School for Computing in Medicine and Life Sciences in University Luebeck Message-ID: <486A476B.5010308@gradschool.uni-luebeck.de> We seek highly skilled and enthusiastic scholars for our newly established Graduate School. We offer 37 interdisciplinary PhD positions in the fields of Neuroengineering, Navigation and Robotics and Computing in Structural and Cell Biology Supervision by leading scientists State-of-the-art research environment Advanced courses for further training Scholarships between 1250? and 1800?. The Graduate School is part of the University of L?beck (Germany) and was established through the Excellence Initiative of the German Research Foundation (DFG). Further information at www.gradschool.uni-luebeck.de. -- Chaoqun Jiang Secretary ================================================================ Graduate School for Computing in Medicine and Life Sciences ---------------------------------------------------------------- Universit?t zu L?beck Ratzeburger Allee 160 D-23538 L?beck Germany Tel: +49 451 500 5670 Fax: +49 451 500 5202 Email: jiang@gradschool.uni-luebeck.de http://www.gradschool.uni-luebeck.de ================================================================ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080701/242492b5/attachment.html From rk at cns.umcn.nl Tue Jul 1 18:30:20 2008 From: rk at cns.umcn.nl (=?ISO-8859-1?Q?Rolf_K=F6tter?=) Date: Wed Jul 2 11:26:40 2008 Subject: [Comp-neuro] Tenure track position at Donders Centre for Neuroscience, Univ. Nijmegen, NL Message-ID: <486A5B9C.9070107@cns.umcn.nl> Dear colleagues: Please see the attached flyer inviting applications for a tenure-track position at the newly founded Donders Centre for Neuroscience, Radboud Univ., Nijmegen, The Netherlands. This is an exciting opportunity for an excellent young researcher with a neuroscience topic of his/her own choice within the scope of the Centre. For more information see www.ru.nl/dcn or contact one of the two people mentioned in the attached flyer. Rolf -- Prof. Dr. Rolf K?tter Donders Centre for Neuroscience, Chair of Section Neurophysiology & Neuroinformatics Department of Cognitive Neuroscience (126) Radboud University Nijmegen Medical Centre Postbus 9101, 6500 HB Nijmegen. Visitors: Geert Grooteplein 21, 6525 EZ Nijmegen, NL phone +31 24 3614248; email rk@cns.umcn.nl fax +31 24 3541435; http://www.neuropi.org -------------- next part -------------- A non-text attachment was scrubbed... Name: Flyer Donders Centre.pdf Type: application/octet-stream Size: 364855 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080701/44a840ea/FlyerDondersCentre-0001.obj From v.steuber at herts.ac.uk Tue Jul 1 19:22:14 2008 From: v.steuber at herts.ac.uk (Volker Steuber) Date: Wed Jul 2 11:26:44 2008 Subject: [Comp-neuro] PhD Studentship in Computational Neuroscience Message-ID: <486A67C6.4080204@herts.ac.uk> PhD Studentship in Computational Neuroscience Science and Technology Research Institute University of Hertfordshire UK Applications are invited for a 3 year PhD Studentship in Computational Neuroscience in the Science and Technology Research Institute at the University of Hertfordshire, UK. The studentship will cover a stipend of ?12,940 per year plus payment of the standard UK student fees. Candidates should be interested in information processing in biologically detailed models of neuronal networks. Our research involves close collaboration with experimentalists in Europe and the USA. More details can be found in these recent publications: Steuber, V., Mittmann, W., Hoebeek, F.E., Silver, R.A., De Zeeuw, C.I., Hausser, M. and De Schutter, E. (2007). Cerebellar LTD and pattern recognition by Purkinje cells. Neuron 54, 121-136. Gleeson, P., Steuber, V. and Silver, R.A. (2007). neuroConstruct: A tool for modeling networks of neurons in 3D space. Neuron 54, 219-35. Calcraft L., Adams R. and Davey N. (2007). Efficient Architectures for Sparsely-Connected High Capacity Associative Memory Models, Connection Science 6, 163-75. Applicants should have good computational and numerical skills and a good first degree in maths, computer science, physics, neuroscience or biology. Previous experience in neuroscience is not required but would be an advantage. The UH Science and Technology Research Institute has been rated as 4 (national excellence with evidence of international excellence) at the last UK university research assessment exercise. It is located in Hatfield in Hertfordshire, just north of London. For informal enquiries contact Dr Volker Steuber (v.steuber@herts.ac.uk ) or Dr Neil Davey (n.davey@herts.ac.uk ). Further information and an application form can be obtained from Mrs Lorraine Nicholls, Research Student Administrator, STRI, Faculty of Engineering and Information Sciences, University of Hertfordshire, College Lane, Hatfield, Herts, AL10 9AB. Tel: +44 1707 286083 Fax: +44 1707 284185 email: l.nicholls@herts.ac.uk. The short-listing process will begin on 25 July 2008. Dr Volker Steuber Senior Lecturer (Research) in Biocomputation Science and Technology Research Institute University of Hertfordshire Hatfield Herts AL10 9AB UK Tel +44 1707 284350 http://homepages.feis.herts.ac.uk/~comqvs/ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080701/6410e00e/attachment.html From nielsen at oist.jp Wed Jul 2 08:46:38 2008 From: nielsen at oist.jp (B. Torben-Nielsen) Date: Wed Jul 2 11:26:46 2008 Subject: [Comp-neuro] Multi-Scale Phenomena in Biology Workshop Message-ID: <486B244E.8090909@oist.jp> Date: November 4, 2008 - November 6, 2008 Location: Okinawa, Japan Okinawa Institute of Science and Technology Promotion Corporation http://www.oist.jp/ "Multi-Scale Phenomena in Biology Workshop" A multitude of biological phenomena are described at multiple levels. What are the commonalities and differences between neuroscience, evolutionary biology, molecular biology and ecology in this regard? How can mathematics help in describing these phenomena? We have speakers from different biological disciplines and from mathematics. Travel scholarships available. We encourage the applications by graduate students and post-docs who's research interests touch these subjects. More information can be obtained from http://www.irp.oist.jp/tenu/multi.html or contact workshop secretariat at multi@oist.jp. Organizers: Robert Sinclair and Klaus Stiefel Confirmed Speakers: Bjorn Engquist Hans Othmer Eric Vanden-Eijnden Keiko Takahashi Dan Rockmore Diego Rasskin-Gutman Klaus M. Stiefel Tony Bell Robert Warner Walter R. Tschinkel Terry Sejnowski -- -------------------------------------------------------------- Ben Torben-Nielsen, Ph.D. student Theoretical and Experimental Neurobiology Unit Okinawa Institute of Science and Technology -------------------------------------------------------------- From bcseet at ieee.org Wed Jul 2 10:35:23 2008 From: bcseet at ieee.org (Boon-Chong Seet) Date: Wed Jul 2 11:26:49 2008 Subject: [Comp-neuro] CFP: 1st International Workshop on Sensor Networks and Ambient Intelligence Message-ID: <007901c8dc1e$94836db0$0701010a@yourbbc104cd11> -------------------------------------------------------------------------------- CFP: 1st International Workshop on Sensor Networks and Ambient Intelligence -------------------------------------------------------------------------------- SeNAmI 2008 1st International Workshop on Sensor Networks and Ambient Intelligence In conjuction with PDCAT'08 http://www.cs.otago.ac.nz/pdcat08 December 1-4, 2008, Dunedin, New Zealand Call for Papers Sensor networks is an enabling technology of Ambient Intelligence. The pervasive nature of unobtrusive sensors distributed in the environment, either embedded or transportable by mobile carriers, enables fine-grain capture of environmental or ambient information that provides the basis of intelligence for higher-order cognitive systems, i.e. systems with capabilities to perceive, reason, learn, and react intelligently to their environment. Such systems in turn are envisioned to have wide ranging applications from intelligent wildlife and building structure monitoring, to humanistic and social endevours such as health and elderly care service provisioning. This workshop aims to bring together researchers from academia and industry to discuss recent research and technology advances in related areas, and from such engagement to foster or stimulate innovations in cross-disciplinary designs and methodologies in the fields of both sensor networks and ambient intelligence. Topics of interest include but are not limited to: - Cognitive wireless sensor networks - Cooperative and distributed sensor localization - Context-aware reasoning and inference for sensor/RFID-based systems - Intelligence support for sensor network management - Sensor networking in heterogeneous wireless environments - Sensor data fusion for ubiquitous embedded computing - Ambient intelligence system architectures, applications, and services - Testbed implementation and experimental trials Manuscript submission Papers reporting original and unpublished research results and experience are solicited. All paper submissions will be handled electronically via EasyChair. Please follow the IEEE Computer Society Press Proceedings Author Guidelines to prepare your papers. Maximum page length of accepted papers will be limited to 6 pages with main body text printed on 10 point font. All accepted papers will be included in conference proceedings of PDCAT'08, which will be published by IEEE Computer Society Press and automatically included in the IEEE Xplore digital library. The proceedings will also be cited by Engineering Information (EI). Selected best papers would be considered for publication in a special issue of the Inderscience International Journal of Autonomous and Adaptive Communication Systems. Important dates Paper submission due : July 21, 2008 (extended) Acceptance notification : August 25, 2008 Camera-ready due : September 1, 2008 Workshop date : TBA For further details, please visit: http://senami08.aut.ac.nz -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080702/0ae4cb70/attachment.html From margret.franke at bccn-berlin.de Wed Jul 2 12:31:34 2008 From: margret.franke at bccn-berlin.de (Margret Franke) Date: Wed Jul 2 13:02:55 2008 Subject: [Comp-neuro] Open position: Teaching Coordinator, Master and PhD Program Computational Neuroscience Message-ID: <486B5906.5000102@bccn-berlin.de> The Bernstein Center for Computational Neuroscience in Berlin has the following open position: Teaching Coordinator Master and PhD Program Computational Neuroscience Starting date: Sep 01, 2008 Applications due: Jul 14, 2008 Applications received after this date might still be considered Tenure: 2 years Remuneration: BAT IIa, part time 30 h Job Description Teaching Coordinator for the international master and PhD program computational neuroscience of the Bernstein Center Berlin. Organisation of the advertising procedures and pre-selection of applicants; consulting of applicants; Participation in the selection of applicants. Management and continuous enhancement of the graduate program; implementation of new teaching contents; quality management of the program; development and distribution of information material for the program; support for the students in general issues like visa, housing, insurance as well as issues concerning the graduate program. Requirements University degree preferably in neuroscience or biology; Very good written and spoken English; Good computer literacy; good organisational skills; team worker Full applications should be sent by Jul 14, 2008 to the contact address given below. Contact Name: Margret Franke Phone: 030-2093-9110 Fax: 030-2093-6771 Email: margret.franke@bccn-berlin Address: Philippstr. 13, Haus 6 10115 Berlin -- Margret Franke Bernstein Center for Computational Neuroscience Institute f. Biology, Humboldt University Philippstr. 13, Haus 6 10115 Berlin phone: (030) 2093-9110 fax: (030) 2093-6771 From vcu at cs.stir.ac.uk Thu Jul 3 16:36:05 2008 From: vcu at cs.stir.ac.uk (Vassilis Cutsuridis) Date: Thu Jul 3 17:54:52 2008 Subject: [Comp-neuro] Perception-action cycle workshop - official euCognition event Message-ID: <4626FDC03DEF4F88B56CCF1661A567F9@cs.ad.stir.ac.uk> Dear colleagues, An euCognition-sponsored workshop on "Adaptive Mechanisms of the Perception-Action Cycle" will be held held in Prague, Czech Republic, in September 6th, in connection with the ICANN conference (www.icann2008.org). The goal of the workshop is to provide an international, interdisciplinary forum on the topic of adaptive mechanisms of the perception-action cycle, with the purpose to advance our understanding of the state-of-the-art on bottom-up and top-down approaches to artificial cognitive systems development. Presentations and papers on perception, attention, memory, learning, decision making, reasoning, conflict resolution, motivation and action will be presented. The manner in which attention is involved (initially to consciously guide the visual and motor processing and then to let it run on automatic until error signals bring attention focus back to the source of the problem and attempt its resolution) are considered highly relevant to the workshop. The perception-action cycle is an important aspect by which to enter a larger domain associated with the construction of autonomous machines. The latter require the perception-action cycle as a basis for development of an embodied system able to learn (by trial and error or observational learning) how to be increasingly effective in the given environment of the machine. More detailed information about the workshop is available at: http://www.icann2008.org/workshop.php http://www.cs.stir.ac.uk/~vcu/ICANN2008.html Best regards, John Taylor, Amir Hussain & Vassilis Cutsuridis (workshop organizers) ---------------------------------------------------------------------------------- Dr. Vassilis Cutsuridis Department of Computing Science and Mathematics University of Stirling Stirling FK9 4LA SCOTLAND Tel: +44 1786 467422 Fax: +44 1786 464551 Email: vcu@cs.stir.ac.uk Web: http://www.cs.stir.ac.uk/~vcu/ -- Academic Excellence at the Heart of Scotland. The University of Stirling is a charity registered in Scotland, number SC 011159. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080703/0dc37528/attachment-0001.html From alexwade at gmail.com Thu Jul 3 19:31:13 2008 From: alexwade at gmail.com (Alex Wade) Date: Fri Jul 4 10:31:25 2008 Subject: [Comp-neuro] Cosyne 2009 Message-ID: <76eaaa9a0807031031q4642b1e1q4de998e7353d377a@mail.gmail.com> ================================================================= Computational and Systems Neuroscience (Cosyne) MAIN MEETING 26 Feb - 1 Mar, 2009 Salt Lake City, Utah WORKSHOPS 2 - 3 Mar, 2009 Snowbird Ski Resort, Utah http://cosyne.org ================================================================= Cosyne is an annual meeting providing an inclusive forum for the exchange of experimental and theoretical approaches to problems in systems neuroscience. The meeting is expected to draw about 350-400 researchers from a wide variety of disciplines. The MAIN MEETING is organized in a single track, and consists of both oral and poster sessions. Some oral presentations are invited (see below), while others are selected based on short submitted abstracts. Poster presentations are also selected from the submitted abstracts. The WORKSHOPS are held in 6-10 parallel sessions, allowing for more in-depth discussion of specialized topics. A Call for Workshop Proposals will be sent out shortly. 2009 CONFIRMED INVITED SPEAKERS: * Richard Axel (Columbia University, USA) * Cori Bargmann (Rockefeller University, USA) * Axel Borst (MPI, Germany) * Jack Gallant (UC Berkeley, USA) * Read Montague (Baylor College of Medicine, USA) * Henry Markram (EPFL, Switzerland) * Earl Miller (MIT, USA) * Jennifer Raymond (Stanford University, USA) * Stephen Scott (Queens University, Canada) * Shihab Shamma (U Maryland, USA) * Joshua Tenenbaum (MIT, USA) * Misha Tsodyks (Weizmann Institute, Israel) ABSTRACT SUBMISSION DEADLINE: 2 Dec 2008 EXECUTIVE COMMITTEE: * Tony Zador (Cold Spring Harbor Laboratory) * Alex Pouget (University of Rochester) * Zach Mainen (Instituto Gulbenkian de Ciencia) ORGANIZING COMMITTEE: * General Chair: Matteo Carandini (University College London) * Program Chair: Maneesh Sahani (University College London) * Workshop Chairs: Adam Kohn (Yeshiva University) and Alex Huk (UT Austin) * Publications Chair: Alex Wade (Smith-Kettlewell Eye Research Institute) ADVISORY BOARD: * Eero Simoncelli (New York University) * Peter Dayan (University College London) * Steven Lisberger (UC San Francisco) * Karel Svoboda (Howard Hughes Medical Institute) From elli.chatzopoulou at incf.org Fri Jul 4 10:03:18 2008 From: elli.chatzopoulou at incf.org (Elli Chatzopoulou) Date: Fri Jul 4 10:31:26 2008 Subject: [Comp-neuro] Accommodation for Neuroinformatics congress Message-ID: <486DD946.6010000@incf.org> *1st INCF Congress of Neuroinformatics: Databasing and Modeling the Brain Stockholm, September 7 - 9, 2008* Register before July 9th to have your hotel room booked through us. Possibility to book your hotel room upon registration available only until July 9th. All pre-booked hotel rooms are in high-standard hotels, across the street from the congress venue. After July 9th, we will not be able to help with this matter and due to the shortage of hotel rooms in central Stockholm, you should expect difficulties in finding other accommodation at a reasonable distance from the venue. Register now and chose your hotel at the same time: https://www.stocon.se/weraform/receive.csp?kgid=816&lang=2 More information about the congress can be found here: http://www.neuroinformatics2008.org/ -- Elli Chatzopoulou, Ph.D. Scientific Information and Public Relations Officer International Neuroinformatics Coordinating Facility Secretariat Karolinska Institutet Nobels v?g 15A SE-171 77 Stockholm Sweden Email: elli.chatzopoulou@incf.org Phone: +46 8 524 87491 Mobile: +46 7 614 87491 Fax: +46 8 524 87150 web: www.incf.org -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080704/1606a9d1/attachment.html From pprodrigues at liaad.up.pt Fri Jul 4 00:16:15 2008 From: pprodrigues at liaad.up.pt (Pedro Pereira Rodrigues) Date: Fri Jul 4 10:31:57 2008 Subject: [Comp-neuro] ACM SAC 2009 - Data Streams Track - Submission Site Now Open Message-ID: <486D4FAF.9040407@liaad.up.pt> *** Apologies for cross-posting *** ACM SAC 2009 - SUBMISSION SITE NOW OPEN! ACM Symposium on Applied Computing The 24nd Annual ACM Symposium on Applied Computing Hilton Hawaiian Village Beach Resort & Spa Waikiki Beach, Honolulu, Hawaii, USA March 8 - 12, 2009 Data Streams Track http://www.liaad.up.pt/~jgama/SAC09/ IMPORTANT DATES Aug 16, 2008: Submission of papers Oct 11, 2008: Notification of acceptance/rejection Oct 25, 2008: Camera-ready copies of accepted papers DATA STREAMS TRACK - CALL FOR PAPERS The rapid development in information science and technology in general and in growth complexity and volume of data in particular have introduced new challenges for the research community. Many sources produce data continuously. Examples include sensor networks, wireless networks, radio frequency identification (RFID), customer click streams, telephone records, multimedia data, scientific data, sets of retail chain transactions, etc. These sources are called data streams. A data stream is an ordered sequence of instances that can be read only once or a small number of times using limited computing and storage capabilities. These sources of data are characterized by being open-ended, flowing at high-speed, and generated by non stationary distributions. TOPICS OF INTEREST We are looking for all possible contributions related to algorithms on data streams. Topics include (but are not restricted) to: Data Stream Models Data Stream Management Systems Data Stream Query Languages Continuous queries and Summarization from Data Streams Sampling Data Streams Single-Pass Algorithms Scalable Algorithms Change Detection Algorithms Clustering on Data Streams Classification and Regression on Data Streams Association Rules on Data Streams Feature Selection on Data Streams Visualization Techniques for Data Streams Evaluation of Data Streams Models Data Stream applications Sensor Networks Real-Time Applications PAPER SUBMISSION GUIDELINES Papers should be submitted in PDF using the SAC 2009 conference management system: http://sac.cs.iupui.edu/sac2009/ The author(s) name(s) and address(es) must NOT appear in the body of the paper, and self-reference should be in the third person. This is to facilitate blind review. Only the title should be shown at the first page without the author's information. The conference proceedings will be published by ACM. Hence, all accepted papers should be submitted in ACM 2-column camera ready format for publication in the symposium proceedings. The maximum number of pages allowed for the final papers is 5 pages (about 4000 words), with the option (at additional expense) to add up to three (3) more pages. There is a set of templates to support the required paper format for a number of document preparation systems at http://www.acm.org/sigs/pubs/proceed/template.html Each submitted paper will be fully refereed and undergo a blind review process by at least three referees. PROGRAM COMMITTEE Jose Avila, University Malaga, Spain Andre Carvalho, University S. Paulo, Brazil Antoine Cornu?jols, Institut National Paris, France Alfredo Cuzzocrea, University of Calabria, Italy Mohamed Gaber, Monash University, Australia Jo?o Gama, University Porto, Portugal Ricard Gavald?, Polytecnic Cataluna, Spain Georges H?brail, Telecom Paris, France Geoff Holmes, University Waikato, New Zealand Eamonn Keogh, University California, United States Ralf Klinkenberg, Rapid-I GmbH, Germany Miroslav Kubat, University Miami, United States Mark Last, University Ben Gorion, Israel Rosa Meo, University of Torino, Italy Pedro Rodrigues, University Porto, Portugal Josep Roure, Polytechnic Cataluna, Spain Elaine Sousa, University S. Paulo, Brazil Eduardo Spinosa, University S. Paulo, Brazil Min Wang, IBM, United States Sean Wang, University Vermon, United States Jiong Yang, Case Western Reserve University, United States Ying Yang, Australian Taxation Office, Australia Philip S. Yu, IBM, United States -------------- next part -------------- A non-text attachment was scrubbed... Name: pprodrigues.vcf Type: text/x-vcard Size: 335 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080703/2ee0aafa/pprodrigues.vcf From hitzler at aifb.uni-karlsruhe.de Fri Jul 4 14:54:18 2008 From: hitzler at aifb.uni-karlsruhe.de (Pascal Hitzler) Date: Fri Jul 4 15:43:20 2008 Subject: [Comp-neuro] CfP Journal Special Issue on Recurrent Neural Networks Message-ID: <486E1D7A.2060300@aifb.uni-karlsruhe.de> Final Call for Papers: Journal Special Issue on == Perspectives and Challenges for Recurrent Neural Networks == Guest Editors: Marco Gori, Barbara Hammer, Pascal Hitzler, Guenther Palm Special issue of the Elsevier Journal of Algorithms in Cognition, Informatics and Logic http://www.elsevier.com/wps/find/journaldescription.cws_home/622851/description = SCOPE = Recurrent neural networks (RNNs) enable flexible machine learning tools which can directly process spatiotemporal and other structured data and which offer a rich dynamic repertoire as time dependent systems. They promise to be efficient signal-processing models which are biologically plausible and optimally suited for a wide range of industrial applications on the one hand, and an explanation of cognitive phenomena of the human brain on the other hand. Despite these facts, however, the design of efficient training methods for RNNs as well as their mathematical investigation with respect to reliable information representation and generalization abilities when dealing with complex data structures is still a challenge. It has led to diverse approaches and architectures including echo and liquid-state-machines, long short term memory, recursive and graph networks, core neuro-symbolic integration, etc. Interestingly, very heterogeneous domains are included, such as logic, chaotic systems, and biological networks. The aim of the special issue is to bring together recent work developed in the field of recurrent information processing, which bridges the gap between different approaches and which sheds some light on canonical solutions or principled problems which occur in the context of recursive information processing when considered across the disciplines. = TOPICS = We particularly encourage submissions connected to the following non-exhaustive list of topics: - new learning paradigms of RNNs such as unsupervised learning or reservoire learning - biologically plausible methods - integration of RNNs and symbolic reasoning - universal approaches for general data structures such as sets or graphs - methods which address the generalization ability of RNNs - challenging applications which have the potential to be benchmark problems - visionary papers concerning the future of RNNs = SUBMISSIONS = Deadline for submissions is 18th of July, 2008. Submissions shall follow the guidelines laid out for the Journal of Algorithms in Cognition, Informatics and Logic, which can be found under . Submissions shall be sent as pdf to Pascal Hitzler, hitzler@aifb.uni-karlsruhe.de = EDITORIAL BOARD = Guilherme da Alencar Barreto, Universidade Federal do Ceara, Brasil Monica Bianchini, University of Siena, Italy Howard Blair, Syracuse University, USA Hendrik Blockeel, KU Leuven, Belgium Mikael Boden, University of Queensland, Australia Matthew Cook, ETH Zuerich, Switzerland Artur d'Avila Garcez, City University London, UK Luc de Raedt, KU Leuven, Belgium Steffen Hoelldobler, TU Dresden, Germany Herbert Jaeger, Jacobs University Bremen, Germany Stefan C. Kremer, University of Guleph, Canada Kai-Uwe Kuehnberger, University of Osnabrueck, Germany Alessio Micheli, University of Pisa, Italy Barak Pearlmutter, NUI Maynooth, Ireland Juergen Schmidhuber, TU Munich, Germany Alessandro Sperduti, University of Padova, Italy Jochen Steil, University of Bielefeld, Germany Peter Tino, University of Bermingham, UK Edmondo Trentin, University of Siena, Italy Thomas Wennekers, University of Plymouth, UK This Call for Papers is available online under http://www.neural-symbolic.org/RNN_CfP.txt -- PD Dr. Pascal Hitzler Institute AIFB, University of Karlsruhe, 76128 Karlsruhe email: hitzler@aifb.uni-karlsruhe.de fax: +49 721 608 6580 web: http://www.pascal-hitzler.de phone: +49 721 608 4751 http://www.neural-symbolic.org From terry at salk.edu Mon Jul 7 00:30:46 2008 From: terry at salk.edu (Terry Sejnowski) Date: Mon Jul 7 10:13:21 2008 Subject: [Comp-neuro] Computational Neurobiology Graduate Program at UCSD In-Reply-To: <4836DC62.40203@bcos.uni-freiburg.de> Message-ID: DEADLINE: DECEMBER 15, 2008 COMPUTATIONAL NEUROBIOLOGY SPECIALIZATION Neurosciences Graduate Training Program - University of California, San Diego http://neurograd.ucsd.edu/doctoral/cnspec.html Overview The Computational Neurobiology Specialization is a new facet of the broader Neuroscience Graduate Program at UCSD. The goal of the specialization is to train the next generation of neuroscientists with the broad range of computational and analytical skills that are essential to understand the organization and function of complex neural systems. The specialization is intended for students with backgrounds in neuroscience, physics, chemistry, biology, psychology, computer science, engineering, and mathematics. The specialization allows Neuroscience students to concentrate on a focused program of rigorous course work in both the theoretical and experimental aspects of computational neuroscience. Students are encouraged to pursue thesis research that includes both an experimental and a computational component, often arranged by the student as a collaboration between two research groups. Upon achievement of degree requirements, students will receive a diploma indicating both their successful completion of the broader Neuroscience Program as well as their specialization in Computational Neurobiology. Themes The program is focused on these major themes relevant for computational neuroscience research: Neurobiology of Neural Systems - the anatomy, physiology, and behavior of systems of neurons, with emphasis on basic phenomenology. Advanced Measurement Tools in Neuroscience - Advanced imaging and recording techniques reflecting the impact of experimental physics on neuroscience. Algorithms for the Analysis of Neural Data - New algorithms and techniques for analyzing data obtained from physiological recording Theoretical Basis for Collective Neural Dynamics - A synthesis of approaches from mathematics and physical sciences as well as biology will be used to explore the collective properties and nonlinear dynamics of neuronal systems. Applications On-line applications: http://neurograd.ucsd.edu/admissions/index.html The deadline for completed application materials, including letters of recommendation, is December 15, 2008. ----- Participating Faculty include: * Henry Abarbanel (Physics): Nonlinear and oscillatory dynamics; modeling central pattern generators in the lobster stomatogastric ganglion. Director, Institute for Nonlinear Systems at UCSD * Thomas Albright (Salk Institute): Motion processing in primate visual cortex; linking single neurons to perception; fMRI in awake, behaving monkeys. Director, Sloan Center for Theoretical Neurobiology * Darwin Berg (Neurobiology): Regulation synaptic components, assembly and localization, function and long-term stability. * Ed Callaway (Salk Institute): Organization and function of neural circuits, visual cortex, genetic & viral methods * Gert Cauwenberghs (Biology): Neuromorphic Engineering; analog VLSI chips; wireless recording and nanoscale instrumentation for neural systems; large-scale cortical modeling. * Andrea Chiba (Cognitive Science): Spatial attention, associative learning, cholinergic, amygdala * EJ Chichilnisky (Salk Institute): Retinal multielectrode recording; neural coding, visual perception. * Garrison Cottrell (Computer Science and Engineering): Dynamical neural network models and learning algorithms * Virginia De Sa (Cognitive Science): Computational basis of perception and learning (both human and machine); multi-sensory integration and contextual influences. * Mark Ellisman (Neurosciences, School of Medicine): High resolution electron and light microscopy; anatomical reconstructions. Director, National Center for Microscopy and Imaging Research * Fred Gage (Salk Institute): Plasticity, neurogenesis, genetics, genomics. Models of neurogenesis in the hippocampus. * Tim Gentner (Psychology): Neuroethology of vocal communication and audition. Models of birdsong learning. * Robert Hecht-Nielsen (Electrical and Computer Engineering): Neural computation and the functional organization of the cerebral cortex. Founder of Hecht-Nielsen Corporation * Steve Hillyard (Neurosciences, School of Medicine): EEG, perception, attention, memory, ERP, SSVEP * Harvey Karten (Neurosciences, School of Medicine): Anatomical, physiological and computational studies of the retina and optic tectum of birds and squirrels * David Kleinfeld (Physics): Active sensation in rats; properties of neuronal assemblies; optical imaging of large-scale activity. * William Kristan (Neurobiology): Computational Neuroethology; functional and developmental studies of the leech nervous system, including studies of the bending reflex and locomotion. Director, Neurosciences Graduate Program at UCSD * Herbert Levine (Physics): Nonlinear dynamics and pattern formation in physical and biological systems, including cardiac dynamics and the growth and form of bacterial colonies * Scott Makeig (Institute for Neural Computation): Analysis of cognitive event-related brain dynamics and fMRI using time-frequency and Independent Component Analysis * Javier Movellan (Institute for Neural Computation): Sensory fusion and learning algorithms for continuous stochastic systems * Mikhael Rabinovich (Institute for Nonlinear Science): Dynamical systems analysis of the stomatogastric ganglion of the lobster and the antenna lobe of insects * Pamela Reinagel (Biology): Sensory and neural coding; natural scene statistics; recordings from the visual system of cats and rodents. * John Reynolds (Salk Institute): Visual attention, cortex, psychophysics, neurophysiology, neural modeling * Massimo Scanziani (Biology): Neural circuits in the somotosensory cortex; physiology of synaptic transmission; inhibitory mechanisms. * Terrence Sejnowski (Salk Institute/Neurobiology): Computational neurobiology; physiological studies of neuronal reliability and synaptic mechanisms. Director, Institute for Neural Computation * Tanya Sharpee (Salk): Statistical physics and information theory approach to understanding sensory processing. Statistical properties of natural auditory and visual environments. * Gabriel Silva (Bioengineering): Functional dynamics of retinal and cortical neural networks, glial signaling physiology, neural engineering * Nicholas Spitzer (Neurobiology): Regulation of ionic channels and neurotransmitters in neurons; effects of electrical activity in developing neurons on neural function. Chair of Neurobiology * Charles Stevens (Salk Institute): Synaptic physiology; theoretical models of neuroanatomical scaling. * Emmanuel Todorov (Cognitive Science): Motor control, stochastic optimal control, sensorimotor loops * Roger Tsien (Chemistry): Second messenger systems in neurons; development of new optical and MRI probes of neuron function, including calcium indicators and caged neurotransmitters ----- From s.schultz at imperial.ac.uk Mon Jul 7 10:22:22 2008 From: s.schultz at imperial.ac.uk (Schultz, Simon R) Date: Mon Jul 7 10:55:55 2008 Subject: [Comp-neuro] PhD studentships at Imperial College Message-ID: <0CE90DDE-16FE-4E34-894D-39724B0DC484@imperial.ac.uk> PhD Studentships Department of Bioengineering Imperial College London The world-class Department of Bioengineering (rated 5* in the Research Assessment Exercise) has available for October 2008 1 BBSRC-funded CASE (industry supported) and 1 standard BBSRC studentships. Eligibility/Duration The BBSRC studentships are open to UK and EU candidates who have been ordinarily resident in the UK for at least 3 years prior to the start of the course. Studentships starting in October 2008 will be for four years. Topics The Departments research interests lie within the following broad themes: Biological and medical imaging Biomechanics and robotics Biophysical and physiological modelling Medical devices and informatics Neuroscience and technology Physiological fluid mechanics Candidates should look at the Departmental Website http://www.imperial.ac.uk/bioengineering to find supervisors and topics of interest within these themes. For the BBSRC studentships the topic needs to be within the remit of the BBSRC and preferably within areas that BBSRC is promoting. Please go to: http://www.bbsrc.ac.uk/science/index.html for further information. Applicants are asked to contact their potential supervisor to discuss putting forward a case ? see email addresses on the Departmental Website. Further information on how to apply can be found on our PhD programme Website: http://www.imperial.ac.uk/bioengineering/courses/phd Deadline: Aug 31st 2008. ------------------- Simon R Schultz Dept of Bioengineering, Imperial College London http://www.imperial.ac.uk/people/s.schultz -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080707/4d8e6419/attachment.html From alexwade at gmail.com Mon Jul 7 23:40:48 2008 From: alexwade at gmail.com (Alex Wade) Date: Tue Jul 8 11:27:37 2008 Subject: [Comp-neuro] Cosyne 2009 - Call for Workshops Message-ID: <76eaaa9a0807071440h42ee72b2s7e38a457a76aede1@mail.gmail.com> --------------------------------------------------------------------------------- Cosyne09 - CALL FOR WORKSHOP PROPOSALS March 2-3, 2009 Snowbird, Utah http://cosyne.org/wiki/Cosyne_09_workshops --------------------------------------------------------------------------------- PROPOSAL DEADLINE: 15 Sept 2008 A series of workshops will be held after the main Cosyne meeting (http://cosyne.org/). The goal is to provide an informal forum for the discussion of important research questions and challenges. Controversial issues, open problems, comparisons of competing approaches, and alternative viewpoints are encouraged. The overarching goal of all workshops should be the integration of empirical and theoretical approaches, in an environment that fosters collegial discussion and debate. Preference will be given to proposals that differ in content, scope, and/or approach from workshops of recent years (examples available at cosyne.org). Relevant topics include, but are not limited to: sensory processing; motor planning and control; multisensory integration; motivation, reward and decision making; learning and memory; adaptation and plasticity; neural coding; neural circuitry and network models; dendritic processing; and methods in computational or systems neuroscience. ________________________________________________________________ WORKSHOP DETAILS: -- There will be 4-8 workshops/day, running in parallel. -- Each workshop is expected to draw between 15 and 80 people. -- The workshops will be split into morning (8:00-11:00 AM) and afternoon (4:30-7:30 PM) sessions. -- Workshops will be held at Snowbird, a ski resort located 30 miles (typically less than an hour) from the Salt Lake City airport. -- Buses from the main conference will be provided. -- Descriptions of previous workshops may be found at http://cosyne.org/wiki/Cosyne_09_workshops ________________________________________________________________ SUBMISSION INSTRUCTIONS: Deadline: September 15th, 2008 Format: plain text only -- please no attachments email to: cosyne09workshops@gmail.com (Alex Huk & Adam Kohn) Proposals should include: - Name(s) and email address(es) of the organizers (no more than 2 organizers per session, please). A primary contact should be designated. -- A title. -- A description of: what the workshop is to address and accomplish, why the topic is of interest, who the targeted group of participants is. -- Names of potential invitees, with indication of which speakers are confirmed. Preference will be given to workshops with the most confirmed speakers. -- Proposed workshop length (1 or 2 days). Most workshops will be limited to a single day. If you think your workshop needs 2 days, please explain why. -- A *brief* resume of the workshop organizer along with a *brief* list of publications (about half a page total). ________________________________________________________________ WORKSHOP ORGANIZERS RESPONSIBILITIES: -- Coordinate workshop participation and content. -- Moderate the discussion. ________________________________________________________________ SUGGESTIONS: Experience has shown that the best discussions during a workshop are those that arise spontaneously. A good way to foster these is to have short talks and long question periods (e.g. 30 + 15 minutes), and have plenty of breaks. Also, when it comes to the number of talks, in the words of Jerry Brown, less is more. We recommend fewer than 10 talks. ________________________________________________________________ WORKSHOP COSTS: Detailed registration costs, etc, will be available at http://cosyne.org/ Please note: Cosyne does NOT provide travel funding for workshop speakers. All workshop participants are expected to pay for workshop registration fees. Participants are encouraged to register early, in order to qualify for discounted registration rates. Cosyne does provide free workshop registration for workshop organizers. ________________________________________________________________ COSYNE 2009 WORKSHOP CHAIRS: Alex Huk (UT Austin), Adam Kohn (Yeshiva U.) QUESTIONS: email: cosyne09workshops@gmail.com (Alex Huk & Adam Kohn) From arno at cerco.ups-tlse.fr Tue Jul 8 12:20:15 2008 From: arno at cerco.ups-tlse.fr (Arnaud Delorme) Date: Tue Jul 8 13:43:08 2008 Subject: [Comp-neuro] Thesis fellowship EEG Neurofeedback to control mind wandering Message-ID: <005D6724-3A50-4468-BA6A-1D2D3D5A4D38@cerco.ups-tlse.fr> Thesis fellowship for 3 years financed by the French ministry of research in Toulouse, France: EEG Neurofeedback to control mind wandering The capacity to concentrate for long period of times is critical both for students and some specific position (air traffic controller, monitor security guard, child care teacher or even Paris taxi driver). Failure to do so can result in life-threatening accident. Monks of various traditions, practicing concentrative meditation, excel at developing a one point focus attention. Over the 3-year course of this project, we intend to 1) record the brain activity of monks in meditation using EEG. 2) train normal non-meditative subjects to produce the same brain rhythms as the monk do. 3) test the concentrative capacities of these subjects before and after training. The institution hosting the project is the University of Toulouse III and the CNRS laboratory CERCO. The CERCO (Center for Brain and Cognition Research) is an internationally recognized laboratory of 18 professional independent CNRS researchers. In 2006 4-year evaluation, this laboratory was ranked first in Neuroscience in France. It is well known for its multidisciplinary approach to cognitive neuroscience with experiments on mice, monkeys and humans, and a wide range of imaging methodologies including modeling. The ideal candidate will have: - a master degree or equivalent in the cognitive or computational domain (no constraint in terms of nationality) - an interest for studying the brain and altered states of consciousness such as those induced by neurofeedback or meditation - a training in mathematics, signal processing and/or computer science - fluency in English. Working knowledge in French is preferred. - capacity to work independently - age limit: 25 (or 30 under special circumstances) Contact: Arnaud Delorme via email before the 31 of August 2008 (arno@cerco.ups-tlse.fr ) (include PDF of a motivation letter, a CV, any publication in english/french and have 2 references send recommendation letter independently by email). http://cerco.ups-tlse.fr/~delorme -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080708/ca53d67a/attachment.html From zhaoyanchang at hotmail.com Wed Jul 9 04:03:06 2008 From: zhaoyanchang at hotmail.com (Yanchang Zhao) Date: Wed Jul 9 11:04:24 2008 Subject: [Comp-neuro] 2nd CFP - DDDM 2008, In conjunction with ICDM'08 Message-ID: ******************************************************************** 2nd Call for Papers - DDDM 2008 The 2nd International Workshop on Domain Driven Data Mining Pisa, Italy, December 15, 2008 In conjunction with IEEE ICDM'08 URL: http://datamining.it.uts.edu.au/dddm08/ ******************************************************************** The Second International Workshop on Domain Driven Data Mining (DDDM 2008) aims to provide a premier forum for sharing findings, knowledge, insight, experience and lessons in tackling potential challenges in discovering actionable knowledge from complex domain problems, promote the interaction of and fill the gap between data mining research and business expectations, and drive a paradigm shift from traditional data-centered hidden pattern mining to domain-driven actionable knowledge discovery. Submission Instructions ----------------------- Paper submissions should be limited to a maximum of 10 pages in the IEEE 2-column format, the same as the camera-ready format (see the IEEE Computer Society Press Proceedings Author Guidelines at http://www.computer.org/portal/pages/cscps/cps/final/icdm06.xml). All papers accepted for the workshop will be included in the ICDM'08 Workshop Proceedings published by the IEEE Computer Society Press. Selected papers from the workshop will be invited for consideration of publication in a planned special issue of IEEE Transactions on Knowledge and Data Engineering (subject to approval from TKDE). Important dates --------------- August 1, 2008: Submission deadline September 15, 2008: Notification of paper acceptance to authors October 7, 2008: Deadline for camera-ready copies December 15, 2008: Workshop day Organizing Committee -------------------- General Chair Philip S. Yu, University of Illinois at Chicago, USA Program Chairs Yanchang Zhao, University of Technology, Sydney, Australia Graham Williams, Australian Taxation Office, Australia Carlos Soares, University of Porto, Portugal Contact ------- Inquiries can be forwarded to dddm08@it.uts.edu.au. _________________________________________________________________ Meet singles near you. Try ninemsn dating now! http://a.ninemsn.com.au/b.aspx?URL=http%3A%2F%2Fdating%2Eninemsn%2Ecom%2Eau%2Fchannel%2Findex%2Easpx%3Ftrackingid%3D1046247&_t=773166080&_r=WL_TAGLINE&_m=EXT From tognoli at ccs.fau.edu Mon Jul 14 17:22:59 2008 From: tognoli at ccs.fau.edu (Emmanuelle TOGNOLI) Date: Thu Jul 17 13:05:00 2008 Subject: [Comp-neuro] Postdoctoral Position: EEG and Behavioral Dynamics of Social Coordination Message-ID: <2387.131.91.30.53.1216048979.squirrel@clifford.ccs.fau.edu> The Human Brain and Behavior Laboratory (HBBL), Center for Complex Systems and Brain Sciences at Florida Atlantic University (FAU) is offering a Postdoctoral Position in the area of Social Neuroscience. The research program aims to study the neural and behavioral mechanisms underlying social coordination: how brain regions couple and decouple both within an individual brain and between brains engaged in social interactions, what factors enhance or degrade the informational coupling between them, or affect the directionality of behavioral influence. The postdoctoral scientist will be highly motivated and will be able to work independently. He/she will also collaborate within an interdisciplinary team of researchers whose expertise spans Neuroscience, Psychology and Physics with the goal of understanding brains and behaviors in terms of complex dynamical systems. The successful applicant will contribute to behavioral and neurobehavioral experiments and to theoretical modeling, in which social interactions are treated as meaningfully coupled dynamical systems. The laboratory is equipped with a high density EEG system (Neuroscan Synamp II) including dual-EEG capability, sound-isolated faraday chamber, Polhemus Isotrack II and Fastrak electrode positioning system, software suites for source reconstruction and multimodal brain imaging (Curry 5). Behavioral facility includes a Northern Digital OPTOTRAK 3010 and mutichannel Analog input/output systems by National Instrument. A 3T Signa Excite MR scanner is also available for extending the work to fMRI and DTI and possibilities for MEG and PET tracer studies are in place. Computationally intensive applications can be directed to a Beowulf cluster maintained by FAU Charles E. Schmidt College of Science. Candidates should have a PhD degree or equivalent, experience or willingness to learn in one or more relevant domains will be considered an advantage: - Preparation and conduct of social, neurobehavioral or neurocognitive experiments - Recording and analysis of EEG or MEG - Digital signal processing and statistical analysis - Programming (Matlab, C, visual basic), - Theoretical modeling, dynamical systems - Excellent writing skills The position will be for one year with the possibility of extension depending on satisfactory progress. Salary will be commensurable with experience. Review of applications will begin immediately and continue until the position is filled. Qualified candidates should send CV and arrange for 3 reference letters via email to: J. A. Scott Kelso, kelso@ccs.fau.edu Emmanuelle Tognoli, tognoli@ccs.fau.edu www.ccs.fau.edu/hbbl.html HBBL, Center for Complex Systems & Brain Sciences, Florida Atlantic University. Boca Raton, FL USA From bcseet at ieee.org Tue Jul 15 06:08:55 2008 From: bcseet at ieee.org (Boon-Chong Seet) Date: Thu Jul 17 13:07:03 2008 Subject: [Comp-neuro] CFP: 1st International Workshop on Sensor Networks and Ambient Intelligence Message-ID: <005a01c8e630$8100a160$790b800a@yourbbc104cd11> -------------------------------------------------------------------------------- CFP: 1st International Workshop on Sensor Networks and Ambient Intelligence -------------------------------------------------------------------------------- SeNAmI 2008 1st International Workshop on Sensor Networks and Ambient Intelligence In conjuction with PDCAT'08 http://www.cs.otago.ac.nz/pdcat08 December 1-4, 2008, Dunedin, New Zealand Call for Papers Sensor networks is an enabling technology of Ambient Intelligence. The pervasive nature of unobtrusive sensors distributed in the environment, either embedded or transportable by mobile carriers, enables fine-grain capture of environmental or ambient information that provides the basis of intelligence for higher-order cognitive systems, i.e. systems with capabilities to perceive, reason, learn, and react intelligently to their environment. Such systems in turn are envisioned to have wide ranging applications from intelligent wildlife and building structure monitoring, to humanistic and social endevours such as health and elderly care service provisioning. This workshop aims to bring together researchers from academia and industry to discuss recent research and technology advances in related areas, and from such engagement to foster or stimulate innovations in cross-disciplinary designs and methodologies in the fields of both sensor networks and ambient intelligence. Topics of interest include but are not limited to: - Cognitive wireless sensor networks - Cooperative and distributed sensor localization - Context-aware reasoning and inference for sensor/RFID-based systems - Intelligence support for sensor network management - Sensor networking in heterogeneous wireless environments - Sensor data fusion for ubiquitous embedded computing - Ambient intelligence system architectures, applications, and services - Testbed implementation and experimental trials Manuscript submission Papers reporting original and unpublished research results and experience are solicited. All paper submissions will be handled electronically via EasyChair. Please follow the IEEE Computer Society Press Proceedings Author Guidelines to prepare your papers. Maximum page length of accepted papers will be limited to 6 pages with main body text printed on 10 point font. All accepted papers will be included in conference proceedings of PDCAT'08, which will be published by IEEE Computer Society Press and automatically included in the IEEE Xplore digital library. The proceedings will also be cited by Engineering Information (EI). Selected best papers would be considered for publication in a special issue of the International Journal of Autonomous and Adaptive Communication Systems (IJAACS). Important dates Paper submission due : July 21, 2008 (extended) Acceptance notification : August 25, 2008 Camera-ready due : September 1, 2008 Workshop date : TBA For further details, please visit: http://senami08.aut.ac.nz From s.schultz at imperial.ac.uk Thu Jul 17 12:31:35 2008 From: s.schultz at imperial.ac.uk (Schultz, Simon R) Date: Thu Jul 17 13:21:04 2008 Subject: [Comp-neuro] Imperial College Junior Research Fellowships Message-ID: <7978DD30-82FC-4B9A-A46B-0361CF7246E1@imperial.ac.uk> I attach details of a recently introduced junior research fellowship scheme at Imperial College - please circulate. We would be happy to support applications in the areas of theoretical/experimental neuroscience or neurotechnology within the Dept. of Bioengineering. -- Imperial College London is delighted to invite applications for our new research career development scheme to support up to 20 research- only Fellowships. The Junior Research Fellowships will give the world's top early-career researchers three years free from teaching and administration plus a competitive salary and laboratory support costs, enabling them to establish and develop their own scientific path. The Fellowships also aim to help scientists make the difficult leap from post-doctoral researcher to lecturer. Fellowships are available across all of Imperial?s core disciplines and can be held in any faculty or the Business school. Each applicant needs to identify a College sponsor who will act as their mentor and help them develop their research. http://www3.imperial.ac.uk/juniorresearchfellowship ------------------- Simon R Schultz Dept of Bioengineering Imperial College London South Kensington Campus Royal School of Mines Building, London SW7 2AZ http://www.imperial.ac.uk/people/s.schultz ------------------- Simon R Schultz Dept of Bioengineering, Imperial College London http://www.imperial.ac.uk/people/s.schultz Phone: 0207 594 1533 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080717/4ca6d1ca/attachment.html From prasad at kitp.ucsb.edu Wed Jul 16 21:09:30 2008 From: prasad at kitp.ucsb.edu (Ila Fiete) Date: Thu Jul 17 13:22:13 2008 Subject: [Comp-neuro] postdoctoral position in computational neuroscience (UT Austin) Message-ID: <487E476A.6020400@kitp.ucsb.edu> Dear Colleagues, I apologize if you receive this post more than once. If relevant, please pass this message on to any enthusiastic young researchers interested in postdoctoral positions in computational neuroscience. ----------------------------------------------------------------------------- Postdoctoral research position in computational neuroscience ----------------------------------------------------------------------------- I'm happy to announce a postdoctoral research position in computational neuroscience in my group at the Center for Learning and Memory, at the University of Texas at Austin. Projects will involve modeling the dynamics of activity, plasticity, and learning in networks that underlie complex behaviors -- including, but not limited to, neural integrators, song production and learning in songbirds, and navigation in ants and rodents. Interactions with the greater neuroscience community are encouraged, and projects will frequently involve collaborations with experimentalists. The initial appointment will be for 1 year with a possibility of extension up to 3 years. Environment: The Center for Learning and Memory is a new center for integrative neuroscience at the University of Texas, Austin, and consists of an energetic mix of junior and senior faculty who use imaging, electrophysiology, molecular biology and biochemistry, genetics, psychology, physics, and computer science techniques to study the brain. The CLM is part of a much larger neuroscience community at UT Austin. UT Austin has excellent programs in engineering, non-linear dynamical systems, and the sciences. For more information, please see http://www.klab.caltech.edu/~ila/ and http://www.clm.utexas.edu/ and http://www.utexas.edu/neuroscience/. To Apply: Applicants should be excited about research and have strong quantitative training in Physics, Mathematics, Computational Neuroscience, Computer Science, or Engineering. A demonstrated dedication to and enthusiasm for research is a must.The ability to program in Matlab or C and some knowledge of neuroscience is desirable but not necessary. If interested, please email me with a copy of your CV or resume and a statement of research interests. Please arrange to have 3 letters of recommendation sent to me by email. I will review applications after receiving all the requested information. Application review will begin immediately, and continue until all positions are filled. Best regards, Ila Fiete From L.Berthouze at sussex.ac.uk Fri Jul 11 13:20:56 2008 From: L.Berthouze at sussex.ac.uk (Luc Berthouze) Date: Thu Jul 17 13:26:49 2008 Subject: [Comp-neuro] Call for Participation: Epigenetic Robotics 2008 Message-ID: ---------------------------------------------------- Apologies for cross-posting ---------------------------------------------------- CALL FOR PARTICIPATION EPIGENETIC ROBOTICS 2008 www.epigenetic-robotics.org Eight International Conference on Epigenetic Robotics - Modeling Cognitive Development in Robotic Systems - ---------------------------------------------------- DATES: July 30-31, 2008 LOCATION: University of Sussex, Brighton, UK EARLY REGISTRATION: Register by July 15 and save! INVITED SPEAKERS: Eva Jablonka (Tel Aviv University, Israel) Epigenetic inheritance in heredity and evolution: A developmental perspective Susan Oyama (John Jay College, New York, USA) Development without roof, without walls, without ?oor Domenico Parisi (CNR, Rome, Italy) How behaviour becomes what it is Claudio Stern (University College London, UK) The magic of gastrulation: from cells to embryo, from molecules to models and back again ---------------------------------------------------- CONFERENCE THEME (for all other information regarding EpiRob'08 please visit www.epigenetic-robotics.org ) In the past 7 years, the Epigenetic Robotics annual conference has established itself as a unique place where original interdisciplinary research from developmental sciences, neuroscience, biology, cognitive robotics, and artificial intelligence is being presented. Psychological theory and empirical evidence is being used to inform epigenetic robotic models, and these models can be used as theoretical tools to make experimental predictions in developmental psychology. As a special feature, this year we are also highlighting a specific organizational theme: evolution and development as related processes of change. The particular focus of this theme is on the dynamic interplay between ontogeny and phylogeny. In other words, how do new abilities and skills that emerge during development influence the path of evolution, and how do subsequent evolutionary changes help to create new developmental trajectories? This is a question that fits well within the mission of epigenetic robotics, as it spans not only a wide range of research areas and academic disciplines (e.g., biology, psychology, AI and machine learning, linguistics, anthropology, etc.) but also a broad spectrum of spatial and temporal scales (e.g., neurons, brains, social communities, cultures, etc.). ---------------------------------------------------- Looking forward to meeting you at EpiRob'08! Dr Luc Berthouze, Senior Lecturer Centre for Computational Neuroscience and Robotics (CCNR) Department of Informatics University of Sussex Brighton BN1 9QH, UK Tel: +44 1273 877206 Fax: +44 1273 877873 From s.li.1 at bham.ac.uk Mon Jul 14 11:27:52 2008 From: s.li.1 at bham.ac.uk (Sheng Li) Date: Thu Jul 17 13:30:08 2008 Subject: [Comp-neuro] Research Fellow in the perception of 3-D shape and surface reflectance, UNIVERSITY OF BIRMINGHAM Message-ID: <468E635F877FE94BBEFFC0309BCA1954E75DBA@psgfs4.adf.bham.ac.uk> SCHOOL OF PSYCHOLOGY UNIVERSITY OF BIRMINGHAM, UK Research Fellow in the perception of 3-D shape and surface reflectance A Wellcome Trust funded position is available to work on a collaborative project between Dr Andrew Welchman (University of Birmingham), Dr Roland Fleming (Max-Planck Institute for Biological Cybernetics, Germany) and Prof. Andrew Blake (Microsoft Research, Cambridge, UK). The successful applicant will combine computational image analysis, psychophysical measurements and modelling to examine the perception of 3-D shape from specular highlights. The work makes use of state-of-the-art rendering techniques and provides the opportunity to work with a high dynamic range display. Research will be conducted within well-equipped labs that incorporate a range of bespoke equipment. The 5* School of Psychology provides an excellent working environment with a pronounced research focus and international expertise in Vision Science, Behavioural Neuroscience and Cognitive Neuroscience. Facilities include an Imaging Centre with integrated equipment for the study of human brain and behaviour (3T scanner, EEG) as well as numerous virtual reality devices and eye trackers. Candidates should hold (or expect to hold) a Ph.D. in Experimental Psychology, Neuroscience, Computer Science, Physics, Mathematics or a related field. Programming skills (e.g. Matlab, C) are essential and experience with simulation, modelling and behavioural testing desirable. Informal enquiries should be directed to Dr Andrew Welchman (A.E.Welchman@bham.ac.uk). Details of salary and application procedures will shortly be available from: www.vacancies.bham.ac.uk/vacancies/ Quoting the reference H47002 Closing date for applications: 24th July 2008 Interviews are anticipated soon after the closing date with the position available from 1st September 2008 From tbosse at few.vu.nl Mon Jul 14 16:15:44 2008 From: tbosse at few.vu.nl (Tibor Bosse) Date: Thu Jul 17 13:32:20 2008 Subject: [Comp-neuro] Second CfP: 2nd Int Workshop on Human Aspects in Ambient Intelligence Message-ID: [Apologies for multiple copies] SECOND INTERNATIONAL WORKSHOP ON HUMAN ASPECTS IN AMBIENT INTELLIGENCE: Agent Technology, Human-Oriented Knowledge and Applications (HAI'08) URL: http://www.few.vu.nl/~treur/HAI08wsCfP.htm Sydney, Australia, December 9, 2008 (Financial support for travelling is available, see below) Workshop at the International Conference on Intelligent Agent Technology (IAT'08) Call for Papers Background Recent developments within Ambient Intelligence and Agent Technology provide new possibilities to contribute to personal care. For example, an intelligent ambient agent in our car may monitor us and warn us when we are falling asleep while driving or take measures when we are too drunk to drive. As another example, an elderly person may wear a device with an ambient agent that monitors his or her wellbeing and generates an action when a dangerous situation is noticed. Such Ambient Intelligence applications can be based on the one hand on possibilities to acquire sensor information about humans and their functioning, but on the other hand, more knowledgeable applications crucially depend on the availability of adequate knowledge for analysis of such information about human functioning. If such knowledge about human functioning is computationally available in intelligent software/hardware devices in the environment, such ambient agents can show more human-like understanding and contribute to personal care based on this understanding. In recent years, scientific areas focusing on human functioning such as cognitive science, psychology, neuroscience and biomedical sciences have made substantial progress in providing an increased insight in the various physical and mental aspects of human functioning. Although much work still remains to be done, models have been developed for a variety of such aspects and the way in which humans (try to) manage or regulate them. From a more biomedical angle, examples of such aspects are (management of) heart functioning, diabetes, eating regulation disorders, and HIV-infection. From a more psychological and social angle, examples are emotion regulation, attention regulation, addiction management, trust management, stress management, and criminal behaviour management. If models of human processes and their management are represented in a formal and computational format, and incorporated in the human environment monitoring the physical and mental state of the human, then such ambient agents are able to perform a more in-depth analysis of the human?s functioning. An ambience is created that has a human-like understanding of humans, based on computationally formalised knowledge from the human-directed disciplines, and that may more effectively affect the state of humans by undertaking in a knowledgeable manner actions that improve their wellbeing and performance. This may concern elderly people and patients, but also humans in highly demanding circumstances or tasks. For example, the workspaces of naval officers may include systems that, among others, track their eye movements and characteristics of incoming stimuli (e.g., airplanes on a radar screen), and use this information in a computational model that is able to estimate where their attention is focussed at. When it turns out that an officer neglects parts of a radar screen, such a system can either indicate this to the person, or arrange on the background that another person or computer system takes care of this neglected part. Aims This workshop series addresses multidisciplinary aspects of Ambient Intelligence and Agent Systems with human-directed disciplines such as psychology, social science, neuroscience and biomedical sciences. The first workshop in the series (HAI'07) took place at the European Conference on Ambient Intelligence (AmI'07), in Darmstadt, Germany, November 2007. The aim of the workshops is to get researchers together from these human-directed disciplines or working on cross connections of Ambient Intelligence with these disciplines. The focus is on the use of knowledge from these disciplines in Ambient Intelligence applications, in order to take care of and support in a knowledgeable manner humans in their daily living in medical, psychological and social respects. The workshop can play an important role, for example, to get modellers in the psychological, neurological, social or biomedical disciplines interested in Ambient Intelligence as a high-potential application area for their models, and, for example, get inspiration for problem areas to be addressed for further developments in their disciplines. From the other side, the workshop may make researchers in Ambient Intelligence, Agent Systems, and Artificial Intelligence more aware of the possibilities to incorporate more substantial knowledge from the psychological, neurological, social and biomedical disciplines in Ambient Intelligence applications. As part of the interaction, specifications may be generated for experiments to be addressed by the human-directed sciences. Some of the areas of interest * human-aware computing * computational modelling of cognitive, neurological, social and biomedical processes for Ambient Intelligence * modelling emotion and mood and their regulation * collecting and analysing histories of behaviour * computational modelling of mindreading, theory of mind * building profiles; user modelling in Ambient Intelligence * sensoring; e.g., tracking physiological states, gaze, body movements, gestures * sensor information integration methods * analysis of sensor information; e.g., voice and skin analysis with respect to emotional states, gesture analysis, heart rate analysis * environmental modelling * situational awareness * model-based reasoning and analysis techniques for Ambient Intelligence * responsive and adaptive systems; machine learning * cognitive agent models * reflective ambient agent architectures * multi-agent system architectures for Ambient Intelligence applications * human interaction with devices * wearable devices for ambient health and wellness monitoring * brain-computer interfacing * analysis and design of applications to care for humans in need of support for physical and mental health; e.g., elderly or psychiatric care, surveillance, penitentiary care, humans in need of strucural medical or psychological care, support for psychotherapeutical/self-help communities * analysis and design of applications to support humans in demanding circumstances and tasks, such as warfare officers, air traffic controllers, crisis and disaster managers, humans in space missions. * evaluation studies * handling aspects of privacy and security; philosophical and ethical aspects Submission and Proceedings Papers can be submitted in the IEEE 2-column format (see the IEEE Computer Society Press Proceedings Author Guidelines, as for the IAT'08 conference). Expected length is from 3 pages (short papers) to 7 pages (regular papers). Double submission is allowed, but inclusion in the proceedings requires that the paper was and is not published elsewhere. For submissions to the main conference IAT'08, it is possible to indicate explicitly that the paper should be considered for the workshop in case of rejection for the main conference. The workshop proceedings will be published by the IEEE Computer Society Press and will be available at the workshop. More submission details are available at the workshop's Website: http://www.few.vu.nl/~treur/HAI08wsCfP.htm Financial Support for Travelling For those presenters at the workshop for whom excessive travelling costs may cause problems, financial support is available. This support may take the form that for a flight ticket above 400 euro, maximally 75% of the total costs of the ticket can be refunded by the workshop organisation (assuming a ticket of reasonable price for the given distance). After acceptance of a paper, this support can be requested for one author of the paper. Registration For every accepted paper at least one author has to register for the WI / IAT-2008 conference. There is no separate workshop registration fee (i.e., only one conference registration covers everything). Important Dates Submission deadline July 30, 2008 Notification September 3, 2008 Camera ready papers September 30, 2008 Workshop December 9, 2008 Coordination Commitee Juan Carlos Augusto (University of Ulster, School of Computing and Mathematics) Tibor Bosse (Vrije Universiteit Amsterdam, Agent Systems Research Group) Cristiano Castelfranchi (CNR Rome, Institute of Cognitive Sciences and Technologies) Diane Cook (Washington State University, USA) Mark Neerincx (TNO Human Factors; Technical University Delft, Man-Machine Interaction) Fariba Sadri (Imperial College, Department of Computing) Jan Treur (contact person, Vrije Universiteit Amsterdam, Agent Systems Research Group) Programme Committee Juan Carlos Augusto (University of Ulster, School of Computing and Mathematics) Marc B?hlen (State University of New York, USA) Tibor Bosse (Vrije Universiteit Amsterdam, Agent Systems Research Group) Antonio Camurri (University of Genoa, InfoMus Lab) Cristiano Castelfranchi (CNR Rome, Institute of Cognitive Sciences and Technologies) Diane Cook (Washington State University, USA) Hao-Hua Chu (National Taiwan University, Ubicomp Lab, Taiwan) Rino Falcone (CNR Rome, Institute of Cognitive Sciences and Technologies) Dirk Heylen (University of Twente, Human Media Interaction) Anthony Jameson (DFKI, Human-Computer Interaction) Judy Kay (University of Sydney, Computer Human Adaptive Interaction, Australia) Peter Leijdekkers (University of Technology Sydney, Mobile Ubiquitous Services & Technologies Group, Australia) Paul Lukowicz (Austrian University for Health Sciences, Medical Informatics and Technology) Silvia Miksch (Danube University Krems, Department of Information and Knowledge Engineering) Jose del Millan (Swiss Federal Institute of Technology in Lausanne EPFL, Research Institute IDIAP, Martigny, Switzerland) Neelam Naikar (Defence Science and Technology Organisation, Centre for Cognitive Work and Safety Analysis, Australia) Tatsuo Nakajima (Waseda University, Distributed and Ubiquitous Computing Lab, Japan) Mark Neerincx (TNO Human Factors; Technical University Delft, Man-Machine Interaction) Toyoaki Nishida (Kyoto University, Department of Intelligence Science and Technology, Japan) Maja Pantic (University of Twente, Human Media Interaction; Imperial College, Department of Computing, Netherlands/UK) Steffen Pauws (Philips Research Europe, Media Interaction Department, Netherlands) Christian Peter (Fraunhofer Institute for Computer Graphics Rostock, Human-Centered Interaction Technologies, Germany) Tomasz M. Rutkowski (RIKEN Brain Science Institute, Laboratory for Advanced Brain Signal Processing, Japan) Fariba Sadri (Imperial College, Department of Computing) Maarten Sierhuis (NASA Ames Research Center, Human-Centered Computing, USA) Elizabeth Sklar (City University of New York, Brooklyn College, Dept of Computer and Information Science) Ron Sun (Rensselaer Polytechnic Institute, Cognitive Science Department) Bruce H. Thomas (University of South Australia Mawson Lakes, Wearable Computer Lab, Australia) Jan Treur (Vrije Universiteit Amsterdam, Agent Systems Research Group) From tom.ferree at gmail.com Thu Jul 10 19:35:02 2008 From: tom.ferree at gmail.com (Thomas Ferree) Date: Thu Jul 17 13:37:07 2008 Subject: [Comp-neuro] Position Announcement: Postdoctoral Research Associate in EEG Signal Processing Message-ID: *Postdoctoral Research Associate in EEG Signal Processing* The University of Texas Southwestern Medical Center at Dallas is seeking outstanding candidates for a postdoctoral research position. Our group uses EEG, fMRI, advanced signal processing and data integration to study neurological diseases. The position is currently funded as part of a large project to study the neurological basis of Gulf War Illness. Some veterans of the first Gulf War (1990-91) were exposed to neurotoxins, and report impairments in attention, semantic processing, memory, and other cognitive functions, as well as motor problems. The goal of our research is to better identify and diagnose this multi-symptom illness, and help guide treatment. A unique aspect of this program is the ability to acquire and analyze literally dozens of kinds of data in each subject, with precisely defined subject populations. As such, this is groundbreaking endeavor in integrative neuroscience. The long-term goal of our group is to adapt this integrative approach to other neurological disorders such as stroke, traumatic brain injury, post-traumatic stress disorder, and attention-deficit/hyperactivity disorder. The laboratory has: 1) 128-channel EEG system (NeuroScan, Inc.) for recording in an acoustic booth, 2) 64-channel EEG system (Brain Products, Inc.) for recording simultaneously with fMRI, 3) 3T MRI scanner (Siemens, Inc.), and 4) mock MR scanner for comparative studies, all devoted solely to our group. The laboratory also plans to purchase a 256-channel electrical impedance tomography system (Electrical Geodesics, Inc.). The successful candidate will work under the supervision of Dr. Thomas Ferree, a physicist with 13 years experience in human EEG and computational neuroscience. The EEG group works closely with experts in MRI, statistics, psychology, psychiatry, neurology, and epidemiology. Minimum qualifications include: 1) BS in physics, mathematics, or equivalent, 2) PhD in physics, mathematics, neuroscience, cognitive science, or related field, 3) excellent programming skills, especially in Matlab, and 4) fluent written and spoken English. The ideal candidate will have expertise in signal processing and statistics, a record of publications in EEG, and interest in bridging computational and cognitive neuroscience. Please send curriculum vitae, cover letter, and the names and contact information of three references to tom.ferree@gmail.com. Review of applications will begin immediately and continue until the position is filled. Ideal start date is late Summer or early Fall 2008. Strong applicants whose materials are received by July 15 may arrange to meet with Dr. Ferree at the Computational Neuroscience Meeting in Portland. The University of Texas Southwestern Medical Center is an Equal Opportunity, Affirmative Action Employer -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080710/aeac356f/attachment-0001.html From tzvi at bu.edu Wed Jul 9 23:38:30 2008 From: tzvi at bu.edu (Uri Eden) Date: Thu Jul 17 13:41:03 2008 Subject: [Comp-neuro] Postdoctoral position at Boston University Center for Neuroscience Message-ID: <48752FD6.30008@bu.edu> Applications are invited for a postdoctoral position with a multidisciplinary research group at the Boston University Center for Neuroscience to study the neural representations of memory encoding and retrieval tasks across multiple data modalities including single unit spiking activity, local field potentials, MEG, EEG, and fMRI. This research will involve working in collaboration with faculty in the Boston University Departments of Mathematics and Statistics and the Center for Memory and Brain in the Department of Psychology to: 1) develop new statistical and data analysis methods to characterize common features of neural coding associated with electrophysiological and imaging data, and 2) characterize the role of theta oscillations in memory encoding and retrieval by developing parametric models both for spiking activity in rat hippocampus in relation to the phase of the theta activity in the LFP and for the theta phase in human EEG/MEG data during behavioral tasks that favor either memory encoding or retrieval. This research will be conducted in collaboration with Professors Howard Eichenbaum, Michael Hasselmo, Chantal Stern, and Uri Eden. The ideal candidate will have a strong quantitative background and a good understanding of basic neuroscience. Candidates with a Ph.D. in any of the following disciplines are encouraged to apply: Neuroscience, Statistics, Engineering, Computer Science, Physics, or any related fields. Experience with signal processing and programming in MATLAB is highly desirable. If interested, please send your curriculum vitae and a brief description of your research interests to Uri Eden at tzvi@bu.edu. From billl at neurosim.downstate.edu Wed Jul 9 14:40:26 2008 From: billl at neurosim.downstate.edu (Bill Lytton) Date: Thu Jul 17 16:43:38 2008 Subject: [Comp-neuro] review announcement: Computer modeling of epilepsy Message-ID: <200807091240.m69CeQSH009104@ru.neurosim.downstate.edu> This review is meant to both introduce epileptologists to computer modeling and to introduce modelers to epilepsy research, an area that is both relatively advanced and relatively well-funded. Lytton WW. Computer modelling of epilepsy. Nat Rev Neurosci. 2008 Jul 2. Epub ahead of print. PMID:18594562 http://www.nature.com/nrn/journal/vaop/ncurrent/abs/nrn2416.html Abstract: Epilepsy is a complex set of disorders that can involve many areas of the cortex, as well as underlying deep-brain systems. The myriad manifestations of seizures, which can be as varied as d?j? vu and olfactory hallucination, can therefore give researchers insights into regional functions and relations. Epilepsy is also complex genetically and pathophysiologically: it involves microscopic (on the scale of ion channels and synaptic proteins), macroscopic (on the scale of brain trauma and rewiring) and intermediate changes in a complex interplay of causality. It has long been recognized that computer modelling will be required to disentangle causality, to better understand seizure spread and to understand and eventually predict treatment efficacy. Over the past few years, substantial progress has been made in modelling epilepsy at levels ranging from the molecular to the socioeconomic. We review these efforts and connect them to the medical goals of understanding and treating the disorder. From bhennon at ucsd.edu Thu Jul 17 16:31:59 2008 From: bhennon at ucsd.edu (Hennon, Brianne) Date: Thu Jul 17 16:43:40 2008 Subject: [Comp-neuro] UC San Diego, Postdoc Position Message-ID: <61B104536E6CE943B33B687BD17D154EC674F8@MCEXCH1.AD.UCSD.EDU> POSITION: Postdoctoral Fellow in Medical Physics LOCATION: University of California, San Diego - Moores Cancer Center Several postdoctoral positions are immediately available in the Center for Advanced Radiotherapy Technologies (CART), Department of Radiation Oncology, University of California, San Diego, for individuals interested in pursuing an academic career in medical physics. Medical physics is a field concerning the application of physics and engineering techniques to solving medical problems. The positions at CART are specifically related to the development of image guidance techniques in cancer radiotherapy. A Ph.D. in physics, engineering, computer science or related disciplines, with strong programming skills, is required. Experience with machine learning will be a plus. Training in clinical medical physics may be provided. Initial appointment will be for one year with potential renewal for a second year. Radiotherapy medical physics is an intellectually challenging yet practical and well compensated profession. It is definitely a career path worthy of exploring for physics and engineering graduates. Interested candidates should e-mail a CV to the contact below. CONTACT: Brianne Hennon Research Administrative Assistant Center for Advanced Radiotherapy Technologies Department of Radiation Oncology University of California, San Diego 3855 Health Sciences Dr. #0843 La Jolla, CA 92093-0843 Tel: (858)822-5036 Fax: (858)822-5568 http://radonc.ucsd.edu/Research/CART -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080717/d251b74a/attachment.html From j.v.stone at sheffield.ac.uk Fri Jul 18 12:09:24 2008 From: j.v.stone at sheffield.ac.uk (Jim Stone) Date: Fri Jul 18 13:14:32 2008 Subject: [Comp-neuro] Key papers in computational neuroscience Message-ID: This is a collection of references obtained in response to a request for key papers from the computational neuroscience community. I have excluded self-citations (but many of those excluded papers actually appear in my own list of key papers below). I have removed the names of respondents, but have left their comments in, as these can be very useful. Many thanks to all those who contributed to this wide-ranging collection. Jim Stone, 18th July 2008. -- JV Stone's key papers: SB Laughlin. A simple coding procedure enhances a neuron's information capacity. Z Naturforsch, 36c:910{912, 1981. See other papers by Laughlin which cover similar material. Lettvin, J.Y., Maturana, H.R., McCulloch, W.S., and Pitts, W.H., What the Frog?s Eye Tells the Frog's Brain, Proc. Inst. Radio Engr. 47:1940-1951, 1959. Ballard, DH, Cortical connections and parallel processing: Structure and function, in Vision, in Brain and cooperative computation, pp 563-621, 1987, Arbib, MA and Hanson AR (Eds). Y Weiss, EP Simoncelli, and EH Adelson. Motion illusions as optimal percepts. Nature Neuroscience, 5(6):598 604, 2002. BA Olshausen and DJ Field. Sparse coding of sensory inputs. Current Opinion in Neurobiology, 14:481 487, 2004. T Poggio, V Torre, and C Koch. Computational vision and regularization theory. Nature, 317:314 319, 1985. AA Stocker and EP Simoncelli. Noise characteristics and prior expectations in human visual speed perception. Nature Neuroscience, 9(4):578 585, 2006. Marr, D., and T. Poggio. <http://cbcl.mit.edu/people/poggio/journals/marr-poggio-science-1976.pdf> Cooperative Computation of Stereo Disparity, Science, 194, 283-287, 1976. Rumelhart, D. E., Hinton, G. E., and Williams, R. J. Learning representations by back-propagating errors. Nature, 323, 533--536. Hinton, G. E. and Nowlan, S. J. How learning can guide evolution. Complex Systems, 1, 495--502. Hinton, G. E. and Plaut, D. C. Using fast weights to deblur old memories. Proceedings of the Ninth Annual Conference of the Cognitive Science Society, Seattle, WA Becker, S. and Hinton, G. E. A self-organizing neural network that discovers surfaces in random-dot stereograms. Nature, 355:6356, 161-163 Ackley, D. H., Hinton, G. E., and Sejnowski, T. J. A learning algorithm for Boltzmann machines. Cognitive Science, 9, 147-169. @article{DURBIN_WILLSHAW_TSP, author ="Durbin, R and Willshaw, D", title = "An analogue approach to the travelling salesman problem using an elastic net method", journal = "Nature", volume = "326", number = "6114", pages = "689-691", month = "", year = "1987" } @article{DOUGLAS_CANONICAL_89, author = "Douglas, RJ and Martin, KAC and Whitteridge, D", title = "A Canonical Microcircuit for Neocortex", journal = "Neural Computation", volume = "1", number = "", pages = "480-488", month = "", year = "1989" } @article{SWINDALE82, author = "Swindale, NV", title = "A model for the formation of orientation columns", journal = "Proceedings Royal Society London B", volume = "215", number = "", pages = "211-230", month = "", year = "1982" } Zohary, E, Shadlen, MN and Newsome, WT (1994). Correlated neuronal discharge rate and its implications for psychophysical performance. Nature 370:140-143. Hopfield's papers (see below). -- Hodgkin and Huxley 1952d (the modeling paper) -- Song and Abbott: Cortical development and remapping through spike timing-dependent plasticity. Neuron 32:339-50, 2001 and Buonomano and Merzevich: Temporal information transformed into a spatial code by a neural network with realistic properties. Science. 1995 Feb 17;267(5200):1028-30. -- Wilson HR, Cowan JD. Excitatory and inhibitory interactions in localized populations of model neurons. Biophys J. 1972 Jan;12(1):1-24. H.B. Barlow, The mechanical mind. Ann. Rev. Neurosci. 13 15-24 (1990) It is about a simple model of consciousness. -- From the cognitive side of computational neuroscience and I recommend: Pouget A, Deneve S, Duhamel JR (2002) A computational perspective on the neural basis of multisensory spatial representations. Nat Rev Neurosci. 3: 741-747. Hamker, F.H., Zirnsak, M., Calow, D., Lappe, M. (2008)?The peri-saccadic perception of objects and space.?PLOS Computational Biology 4(2):e21 Olshausen BA, Field DJ. 1996. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381:607-9. -- I was really influenced by @article{Atick92, Author = {Atick, Joseph J.}, Journal = {Network: {C}omputation in {N}eural {S}ystems}, Number = {2}, Pages = {213--52}, Title = {Could {I}nformation {T}heory {P}rovide an {E}cological {T}heory of {S}ensory {P}rocessing?}, Volume = {3}, Year = {1992}} which is a review paper rather related to the seminal papers from Barlow and Marr. -- Wilson HR, Cowan JD. Excitatory and inhibitory interactions in localized populations of model neurons. Biophys J. 1972 Jan;12(1):1-24. -- Wiring Optimization Dmitri B. Chklovskii Traub's CA1 model/Pinsky Rinzel 2 compartmental models Erik De Schutter's Purkinje cell models Henry Markram's Cortical Models Rolls & Treves - Hippocampal Network Polsky & Mel - 2layer pyramidal cell model Terry Sejnowski - Synapse, modeldb Upinder S Bhalla - Million Synapses / Bistable systems -- These papers introduced accurate models of calcium dynamics and neuromodulatory effects on ion channel activity. Bhalla US, Iyengar R. Emergent properties of networks of biological signaling pathways. Science. 1999 Jan 15;283(5400):381-7. Zador A, Koch C, Brown TH. Biophysical model of a Hebbian synapse. Proc Natl Acad Sci U S A. 1990 Sep;87(17):6718-22. Holmes WR, Levy WB.Abstract Insights into associative long-term potentiation from computational models of NMDA receptor-mediated calcium influx and intracellular calcium concentration changes. J Neurophysiol. 1990 May;63(5):1148-68. -- There are two theoretical papers which, in my opinion, have had a strong influence on the way we think about synaptic transmission and short term plasticity today: A W Liley and K A North. An electrical investigation of effects of repetitive stimulation on mammalian neuromuscular junction. J Neurophysiol, 16(5):509  527, Sep 1953. W J Betz. Depression of transmitter release at the neuromuscular junction of the frog. J Physiol, 206(3):629 644, 1970. These were, of course, published before the term "computation neuroscience" was used. The first proposed a mathematical model for vesicle pool depletion, which is still in use today. The second was the first to extend this with the release probability as a dynamic variable. These ideas were then further popularised by these classic papers: L F Abbott, J A Varela, K Sen, and S B Nelson. Synaptic depression and cortical gain control. Science, 275(5297):220 224, Jan 1997. M V Tsodyks and H Markram. The neural code between neocortical pyramidal neurons depends on neurotransmitter release probability. Proc Natl Acad Sci U S A, 94(2):719 723, Jan 1997. What I found have during my collaborations with biologists was that not so much the precise mathematical formulation, but the very basic ideas and concepts explored in these papers have made a strong impact in the whole field, and have certainly cleared the way for numerous further theoretical studies. Another paper I have come across just recently which I would consider as rather important and useful is this: J J Hopfield and A V M Herz. Rapid Local Synchronization of Action Potentials: Toward Computation with Coupled Integrate-and-Fire Neurons. Proc Natl Acad Sci U S A, 92(15): 6655-6662, Jul 1995. Cited more than 150 times, it contains some strong results regarding the behaviour of recurrent networks, and also anticipates a number of results shown more recently. -- Here is my top 12 papers, in chronological order. I have gone for ones that make my science heart sing, that introduce a big idea, useful tool, connect experiment and theory in a satisfying way, or are an example of work on a topic that has been mysteriously under-represented. I have tried to briefly qualify why they could be thought of as classic by the wider community. 1) Willshaw and von der Malsburg (1979). Future hot topic: modelling development Excellent interaction between theory and experiment - predicted ephrins and eph receptors. http://www.jstor.org/stable/pdfplus/2418226.pdf 2) Laughlin (1981) Z. Naturforsch. C 36:910-2 Big idea: coding matches stimulus statistics. http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&TermToSearch=7303823 3) Srinivasan et al. (1982) Proc. Roy. Soc. B 216(1205):427-59 Excellent interaction between theory and experiment: predicts responses of first order visual interneurons if they exploit spatial and temporal correlations to reduce redundancy. http://www.kyb.tuebingen.mpg.de/bethgegroup/teaching/ws0708_sem_retina_w hitening/Srinivasan_et_al_1982.pdf 4) Buchsbaum and Gottschalk (1983). Proc. R. Soc. B 220:89-113 Excellent interaction between theory and experiment: uses PCA to accurately calculate the colour channels that maximise information transmission. Deserves to be more widely known. http://www.jstor.org/stable/pdfplus/35873.pdf 5) Bialek et al. (1991) Science Useful application for theorist: neat method for calculating stimulus filters in the response. http://www2.hawaii.edu/~sstill/neural_code_91.pdf 6) Treves and Rolls (1992) Hippocampus 2(2):189-99 Excellent interaction between theory and experiment: identified the function of the dentate gyrus in the hippocampus, and matched network organisation to function far more successfully that Marr. http://www3.interscience.wiley.com/cgi-bin/fulltext/109711333/PDFSTART 7) Van Hateren (1992) J. Comp Phys. A 171:157-170 Excellent interaction between theory and experiment: predicts visual spatiotemporal receptive fields of cells connected to photoreceptors in the fly so as to maximise information about natural images from first principles, with stunning success. http://www.springerlink.com/content/h4681x344j378229/fulltext.pdf 8) Wolpert et al. (1995) Science 269(5232):1880-2 Big idea: internal models and the use of priors. http://keck.ucsf.edu/~houde/sensorimotor_jc/DMWolpert95a.pdf 9) Zemel et al. (1998) Neur. Comp. 10(2):403-30 Big idea: neurons encode distributions, not single values http://www.gatsby.ucl.ac.uk/~dayan/papers/zdp98.pdf 10) Van Rossum et al. (2000) J. Neuro. 20(23):8812-21 Excellent interaction between theory and experiment: Simple application of Fokker-Planck equation physics to explain functional consequences to the network of cellular level experimental data. http://www.jneurosci.org/cgi/reprint/20/23/8812.pdf 11) Brunel (2000) J. Comp. Neuro 8:183-208 Useful application for theorist: calculations of the population activity of a network of integrate-and-fire neurons. http://www.springerlink.com/content/u446l5722lp03677/fulltext.pdf 12) Schreiber (2000) Physical Review Letters 85(2):461-64 Future hot topic: Current best method to infer causal relationships between neurons using information theory. http://prola.aps.org/pdf/PRL/v85/i2/p461_1 -- Here are the most important papers in 3 subjects, plasticity and simple neuron models and network dynamics Of course, there are other categories in Computational neuroscience (detailed neuron model, cortex modeling, vision, audition etc) on which others will report. 1) In plasticity: Hebb, 1949 (book) Bienenstock, Cooper Munro, J. Neurosci.1982 (BCM rule) Kohonen Neural Networks1993 (Kohonen algo in comp neuro perspective other papers of him would also do) Hopfield, PNAS, 1982 (Hopfield model) Amit Gutfreund Sompolinksy, Phys Rev A, 1985 (Analysis of Hopfield model) Linsker PNAS, 1986 (emergence of field) MacKay and Miller 1990 Neural Comput. (analysis of Linskers rule) Miller and MacKay 1994 Neural Comput. the role of constraints Gerstner et al, Nature 1996 (first paper on STDP) Kempter et al. Phys Rev E, 1999 (first analysis of STDP) Lisman, PNAS, 1999 (first model of plasticity based on calcium dynamics) Song Miller Abbott, Nat. Neurosci, 2000 (popular paper on STDP) Rossum et al. 2000, J. Neuroscie (STDP with soft bounds for the weights) Fusi, Biological Cybernetics, 2002 (some general problems of Hebbian rules - nice review of work of Fusi)_ Shouval et al., PNAS, 2002 (calcium model of plasticity) Senn Tsodyks, Markram, Neural. comp. 2001 (STDP algorithm) Fusi, Drew, Abbott Nat. Neuroscience 2005 (Cascade model) Toyoizumi et al. PNAS 2005 (BCM rule for spiking neuron also optimized information) 2) In simplified neuron models Lapicque 07 (often cited as first integrate-and-fire model, even though it does not show reset) FitzHugh 1961, Biophys. Journal (2-dim neuron model) Stein 1967, Biophys. Journal (some models of neural variability - integrate-and-fire model with noise) Ermentrout 1996, Neural Comput., Canonical type I model, quadratic integrate-and-fire Kistler et al. 1997, Neural Computation (systematic reduction to a threshold model/Spike Response Model) Latham 2000, J. Neurophys. quadratic integrate-and-fire Izhikevich 2003, IEEE, 2-dim. neuron model Fourcaud et al. 2003, J. Neurosci. exp. integrate-and-fire model Jolivet et al. 2006, J. comput. Neurosci. -- spiking in real neurons can be explained by threshold models Badel et al. 2008, J. Neurophysiol. -- real neurons are exponential integrate-and-fire models, this is a very recent paper, but it is really important for the discussion of simple neuron models 3) Network dynamics Wilson and Cowan, 1972 Amari 1974 Brunel and Hakim, 1999 Neural Computation Gerstner 2000 Neural Computation Brunel 2000 Comput. Neurosci -- Finally, I am attaching a list of great papers. If I were trying to get outsiders excited, I'd definitely use the Andy Schwartz paper on neural prosthetics. Also think I would do Olshausen & Field as it really kicked people off on thinking about natural images. The Hopfield paper is the greatest of the bunch but is likely too old for what you're looking for. Spike-timing-dependent plasticity is a hot topic and I think carries on a great tradition of computational neuroscientists connecting cellular plasticity to larger network functions; and I think Peter Dayan (and Montague's in the original paper) work is some of the first that really puts a framework in place for thinking about neuromodulators. But they're all great, and I tried to hit many different contributions (maybe this is the greatest message--that computational neuroscience pervades so many fields from single-neuron computation to neuromodulators to models of memory). 1. Montague PR, Dayan P, Sejnowski TJ A framework for mesencephalic dopamine systems based on predictive Hebbian learning. J Neurosci. 1996 Mar 1;16(5):1936-47. Abstract: We develop a theoretical framework that shows how mesencephalic dopamine systems could distribute to their targets a signal that represents information about future expectations. In particular, we show how activity in the cerebral cortex can make predictions about future receipt of reward and how fluctuations in the activity levels of neurons in diffuse dopamine systems above and below baseline levels would represent errors in these predictions that are delivered to cortical and subcortical targets. We present a model for how such errors could be constructed in a real brain that is consistent with physiological results for a subset of dopaminergic neurons located in the ventral tegmental area and surrounding dopaminergic neurons. The theory also makes testable predictions about human choice behavior on a simple decision-making task. Furthermore, we show that, through a simple influence on synaptic plasticity, fluctuations in dopamine release can act to change the predictions in an appropriate manner. This paper is the first of a series of papers setting up a framework for how mesencephalic dopamine neurons represent reward and can serve as the basis for temporal difference -based reward learning in which the reward is offered at a delayed time. *2. Strong, S., Koberle, R., de Ruyter van Steveninck, R. and Bialek, W. 1998. Entropy and information in neural spike trains, Physical Review Letters 80: 197-200. Abstract. The nervous system represents time dependent signals in sequences of discrete, identical action potentials or spikes; information is carried only in the spike arrival times. We show how to quantify this information, in bits, free from any assumptions about which features of the spike train or input signal are most important, and we apply this approach to the analysis of experiments on a motion sensitive neuron in the fly visual system. This neuron transmits information about the visual stimulus at rates of up to 90 bits/s, within a factor of 2 of the physical limit set by the entropy of the spike train itself. This paper ushered in a new set of techniques for characterizing spike trains using the methods of information theory, and also illustrated that there was information on much smaller time scales (~a couple ms) than had typically been assumed previously. 3a. Abbott LF, Varela JA, Sen K, Nelson SB. Synaptic depression and cortical gain control.Science. 1997 Jan 10;275(5297):220-4 Abstract. Cortical neurons receive synaptic inputs from thousands of afferents that fire action potentials at rates ranging from less than 1 hertz to more than 200 hertz. Both the number of afferents and their large dynamic range can mask changes in the spatial and temporal pattern of synaptic activity, limiting the ability of a cortical neuron to respond to its inputs. Modeling work based on experimental measurements indicates that short-term depression of intracortical synapses provides a dynamic gain-control mechanism that allows equal percentage rate changes on rapidly and slowly firing afferents to produce equal postsynaptic responses. Unlike inhibitory and adaptive mechanisms that reduce responsiveness to all inputs, synaptic depression is input-specific, leading to a dramatic increase in the sensitivity of a neuron to subtle changes in the firing patterns of its afferents. -AND- 3b. Markram H, Tsodyks M. Redistribution of synaptic efficacy between neocortical pyramidal neurons. Nature. 1996 Aug 29;382(6594):807-10. Abstract. Experience-dependent potentiation and depression of synaptic strength has been proposed to subserve learning and memory by changing the gain of signals conveyed between neurons. Here we examine synaptic plasticity between individual neocortical layer-5 pyramidal neurons. We show that an increase in the synaptic response, induced by pairing action-potential activity in pre- and postsynaptic neurons, was only observed when synaptic input occurred at low frequencies. This frequency-dependent increase in synaptic responses arises because of a redistribution of the available synaptic efficacy and not because of an increase in the efficacy. Redistribution of synaptic efficacy could represent a mechanism to change the content, rather than the gain, of signals conveyed between neurons. These 2 papers connected short-term synaptic plasticity to important computational implications. 4a. Hopfield JJ. Neural networks and physical systems with emergent collective computational abilities. Proc Natl Acad Sci U S A. 1982 Apr;79(8):2554-8 Abstract. Computational properties of use of biological organisms or to the construction of computers can emerge as collective properties of systems having a large number of simple equivalent components (or neurons). The physical meaning of content-addressable memory is described by an appropriate phase space flow of the state of a system. A model of such a system is given, based on aspects of neurobiology but readily adapted to integrated circuits. The collective properties of this model produce a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size. The algorithm for the time evolution of the state of the system is based on asynchronous parallel processing. Additional emergent collective properties include some capacity for generalization, familiarity recognition, categorization, error correction, and time sequence retention. The collective properties are only weakly sensitive to details of the modeling or the failure of individual devices. This classic paper illustrated the idea of attractor models and a correspondence with energy surfaces. It is now universally permeates discussions of long-term memory storage in networks, especially in the hippocampus. It was followed more recently by the article below, which expanded the idea of attractor models to continuous attractors this now is the framework for discussion of many networks storing short-term memories (the other set of models being the so-called ring models but i am not sure of the original reference for those). 4b. Seung HS. How the brain keeps the eyes still. Proc Natl Acad Sci U S A. 1996 Nov 12;93(23):13339-44. Abstract. The brain can hold the eyes still because it stores a memory of eye position. The brain's memory of horizontal eye position appears to be represented by persistent neural activity in a network known as the neural integrator, which is localized in the brainstem and cerebellum. Existing experimental data are reinterpreted as evidence for an "attractor hypothesis" that the persistent patterns of activity observed in this network form an attractive line of fixed points in its state space. Line attractor dynamics can be produced in linear or nonlinear neural networks by learning mechanisms that precisely tune positive feedback. 5a. Song S, Miller KD, Abbott LF. Competitive Hebbian learning through spike-timing-dependent synaptic plasticity. Nat Neurosci. 2000 Sep;3(9):919-26. Abstract. Hebbian models of development and learning require both activity-dependent synaptic plasticity and a mechanism that induces competition between different synapses. One form of experimentally observed long-term synaptic plasticity, which we call spike-timing-dependent plasticity (STDP), depends on the relative timing of pre- and postsynaptic action potentials. In modeling studies, we find that this form of synaptic modification can automatically balance synaptic strengths to make postsynaptic firing irregular but more sensitive to presynaptic spike timing. It has been argued that neurons in vivo operate in such a balanced regime. Synapses modifiable by STDP compete for control of the timing of postsynaptic action potentials. Inputs that fire the postsynaptic neuron with short latency or that act in correlated groups are able to compete most successfully and develop strong synapses, while synapses of longer-latency or less-effective inputs are weakened. -AND- 5b. Song S, Abbott LF.Neuron. Cortical development and remapping through spike timing-dependent plasticity. 2001 Oct 25;32(2):339-50 Abstract. Long-term modification of synaptic efficacy can depend on the timing of pre- and postsynaptic action potentials. In model studies, such spike timing-dependent plasticity (STDP) introduces the desirable features of competition among synapses and regulation of postsynaptic firing characteristics. STDP strengthens synapses that receive correlated input, which can lead to the formation of stimulus-selective columns and the development, refinement, and maintenance of selectivity maps in network models. The temporal asymmetry of STDP suppresses strong destabilizing self-excitatory loops and allows a group of neurons that become selective early in development to direct other neurons to become similarly selective. STDP, acting alone without further hypothetical global constraints or additional forms of plasticity, can also reproduce the remapping seen in adult cortex following afferent lesions. The papers above have been seminal in illustrating the implications for learning of spike-timing-dependent synaptic plasticity 6. Polsky A, Mel BW, Schiller J. Nat Neurosci. 2004 Jun;7(6):621-7. Epub 2004 May 23. Computational subunits in thin dendrites of pyramidal cells. Abstract. The thin basal and oblique dendrites of cortical pyramidal neurons receive most of the synaptic inputs from other cells, but their integrative properties remain uncertain. Previous studies have most often reported global linear or sublinear summation. An alternative view, supported by biophysical modeling studies, holds that thin dendrites provide a layer of independent computational 'subunits' that sigmoidally modulate their inputs prior to global summation. To distinguish these possibilities, we combined confocal imaging and dual-site focal synaptic stimulation of identified thin dendrites in rat neocortical pyramidal neurons. We found that nearby inputs on the same branch summed sigmoidally, whereas widely separated inputs or inputs to different branches summed linearly. This strong spatial compartmentalization effect is incompatible with a global summation rule and provides the first experimental support for a two-layer 'neural network' model of pyramidal neuron thin-branch integration. Our findings could have important implications for the computing and memory-related functions of cortical tissue. This paper, as well as previous theoretical work, suggests that dendrites might enable single neurons to behave as feedforward neural networks. 7. Medina JF, Nores WL, Mauk MD. Nature. 2002 Mar 21;416(6878):330-3. Inhibition of climbing fibres is a signal for the extinction of conditioned eyelid responses. Abstract. A fundamental tenet of cerebellar learning theories asserts that climbing fibre afferents from the inferior olive provide a teaching signal that promotes the gradual adaptation of movements. Data from several forms of motor learning provide support for this tenet. In pavlovian eyelid conditioning, for example, where a tone is repeatedly paired with a reinforcing unconditioned stimulus like periorbital stimulation, the unconditioned stimulus promotes acquisition of conditioned eyelid responses by activating climbing fibres. Climbing fibre activity elicited by an unconditioned stimulus is inhibited during the expression of conditioned responses-consistent with the inhibitory projection from the cerebellum to inferior olive. Here, we show that inhibition of climbing fibres serves as a teaching signal for extinction, where learning not to respond is signalled by presenting a tone without the unconditioned stimulus. We used reversible infusion of synaptic receptor antagonists to show that blocking inhibitory input to the climbing fibres prevents extinction of the conditioned response, whereas blocking excitatory input induces extinction. These results, combined with analysis of climbing fibre activity in a computer simulation of the cerebellar-olivary system, suggest that transient inhibition of climbing fibres below their background level is the signal that drives extinction. This is one of several computational studies by Mauk and collaborators that are enhancing our knowledge of cerebellar processing (also see similar papers by Raymond & Lisberger applied to the VOR). 8. Olshausen BA, Field DJ. Nature. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. 1996 Jun 13;381(6583):607-9 Abstract.The receptive fields of simple cells in mammalian primary visual cortex can be characterized as being spatially localized, oriented and bandpass (selective to structure at different spatial scales), comparable to the basis functions of wavelet transforms. One approach to understanding such response properties of visual neurons has been to consider their relationship to the statistical structure of natural images in terms of efficient coding. Along these lines, a number of studies have attempted to train unsupervised learning algorithms on natural images in the hope of developing receptive fields with similar properties, but none has succeeded in producing a full set that spans the image space and contains all three of the above properties. Here we investigate the proposal that a coding strategy that maximizes sparseness is sufficient to account for these properties. We show that a learning algorithm that attempts to find sparse linear codes for natural scenes will develop a complete family of localized, oriented, bandpass receptive fields, similar to those found in the primary visual cortex. The resulting sparse image code provides a more efficient representation for later stages of processing because it possesses a higher degree of statistical independence among its outputs. This now classic study suggests how the statistical structure of natural images may determine the response properties of V1 cells, and set the stage for many later studies discussing the concept of sparse coding of images. 9. Taylor DM, Tillery SI, Schwartz AB. Direct cortical control of 3D neuroprosthetic devices. Science. 2002 Jun 7;296(5574):1829-32. Abstract. Three-dimensional (3D) movement of neuroprosthetic devices can be controlled by the activity of cortical neurons when appropriate algorithms are used to decode intended movement in real time. Previous studies assumed that neurons maintain fixed tuning properties, and the studies used subjects who were unaware of the movements predicted by their recorded units. In this study, subjects had real-time visual feedback of their brain-controlled trajectories. Cell tuning properties changed when used for brain-controlled movements. By using control algorithms that track these changes, subjects made long sequences of 3D movements using far fewer cortical units than expected. Daily practice improved movement accuracy and the directional tuning of these units. This represents some of the seminal work decoding cortical activity to control neural prosthetics. 10. Van Vreeswijk C, Abbott LF, Ermentrout GB. When inhibition not excitation synchronizes neural firing. J Comput Neurosci. 1994 Dec;1(4):313-21. Abstract. Excitatory and inhibitory synaptic coupling can have counter-intuitive effects on the synchronization of neuronal firing. While it might appear that excitatory coupling would lead to synchronization, we show that frequently inhibition rather than excitation synchronizes firing. We study two identical neurons described by integrate-and-fire models, general phase-coupled models or the Hodgkin-Huxley model with mutual, non-instantaneous excitatory or inhibitory synapses between them. We find that if the rise time of the synapse is longer than the duration of an action potential, inhibition not excitation leads to synchronized firing. If I were to update, I think I would add papers from: 1) Neuroeconomics & Reinforcement Learning-- in addition to the seminal work by Dayan & Schultz (already in attached), perhaps Loewenstein/Seung paper on matching behavior as a generic consequence of correlational learning rules. 2) Bayesian networks -- maybe Ma, Beck, Latham, Pouget or others on idea that the brain may encode & compute with probabilities -- END OF FILE -- -- Dr Jim Stone, Psychology Department, Sheffield University, Sheffield, S10 2TP, UK. Tel: 0114 2226522. http://jim-stone.staff.shef.ac.uk/ From j.v.stone at sheffield.ac.uk Fri Jul 18 14:54:48 2008 From: j.v.stone at sheffield.ac.uk (Jim Stone) Date: Fri Jul 18 15:25:37 2008 Subject: [Comp-neuro] pdf format of: Key papers in computational neuroscience Message-ID: Apparently some oddity in the text I included in my previous email made it difficult for some to read. Please find attached pdf of same. Jim Stone -- Dr Jim Stone, Psychology Department, Sheffield University, Sheffield, S10 2TP, UK. Tel: 0114 2226522. http://jim-stone.staff.shef.ac.uk/ -------------- next part -------------- A non-text attachment was scrubbed... Name: KeyCompNeuroPapers.pdf Type: application/octet-stream Size: 48179 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080718/4198dbd0/KeyCompNeuroPapers-0001.obj From nurban at cmu.edu Fri Jul 18 14:41:22 2008 From: nurban at cmu.edu (Nathan Urban) Date: Fri Jul 18 15:25:51 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <20080718111439.4E18A8FEE3C@neuroinf.org> References: <20080718111439.4E18A8FEE3C@neuroinf.org> Message-ID: <48808F72.80501@cmu.edu> Review announcement This review describes a constructive role for noise in synchronizing populations of neurons and should be of interest to computaional neurosciuentists. Trends Neurosci. 2008 Jul 4. [Epub ahead of print] Reliability, synchrony and noise. Ermentrout GB, Gal?n RF, Urban NN. The brain is noisy. Neurons receive tens of thousands of highly fluctuating inputs and generate spike trains that appear highly irregular. Much of this activity is spontaneous - uncoupled to overt stimuli or motor outputs - leading to questions about the functional impact of this noise. Although noise is most often thought of as disrupting patterned activity and interfering with the encoding of stimuli, recent theoretical and experimental work has shown that noise can play a constructive role - leading to increased reliability or regularity of neuronal firing in single neurons and across populations. These results raise fundamental questions about how noise can influence neural function and computation. PMID: 18603311 [PubMed - as supplied by publisher] http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_user=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_version=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 From alexey at math.iupui.edu Fri Jul 18 15:31:56 2008 From: alexey at math.iupui.edu (Alexey Kuznetsov) Date: Fri Jul 18 16:11:24 2008 Subject: [Comp-neuro] Post-Doctoral Position in Mathematical Biology Message-ID: Indiana University-Purdue University Indianapolis Post-Doctoral Position in Mathematical Biology The IUPUI Department of Mathematical Sciences and the Center for Mathematical Biosciences invite applications for a one-year postdoctoral position in mathematical and computational biology, beginning September 1, 2008 (or later). This position may be renewed for an additional year based upon the availability of funds. We are located in Indianapolis on the IUPUI Health Sciences campus, the focal point of the Indiana Life Sciences Initiative, which provides an excellent environment for collaborative research in many areas of quantitative biosciences. The postdoctoral fellow will participate in an ongoing project on the exploration of dynamical properties and signal processing performed by the brain. The project has strong mathematical and biological components, offering interaction with experimental collaborators. Please visit www.math.iupui.edu/~alexey/research.html for more information. Applicants should have a Ph.D in Mathematics, Physics, or a related field, as well as a background in Dynamical Systems and programming experience. Working knowledge in Computational Neuroscience is also desired. To apply, please send your CV, list of publications, research statement, and contact information for three references to Prof. Alexey Kuznetsov via e-mail: akuznetsov@math.iupui.edu, or by regular mail: 402 N. Blackford Street, LD 270, Indianapolis, IN 46202. IUPUI is an EEO/AA Employer, M/F/D. Alexey Kuznetsov Assistant Professor Dept. of Mathematical Sciences, IUPUI Science Building, LD 270, 402 N. Blackford St., Indianapolis, IN 46202-3216 From bower at uthscsa.edu Fri Jul 18 15:56:07 2008 From: bower at uthscsa.edu (jim bower) Date: Fri Jul 18 16:20:48 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <48808F72.80501@cmu.edu> References: <20080718111439.4E18A8FEE3C@neuroinf.org><48808F72.80501@cmu.edu> Message-ID: <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry> Haven't done this in a long time. But who says neurons are noisy? From the point of view of information theory, why isn't the apperance of noise expected in a highly optimized coding scheme? And why isn't synchrony to be avoided as redundency. Engineers avoid it, why shouldn't evolution. Just curious. Jim bower Sent via BlackBerry by AT&T -----Original Message----- From: Nathan Urban Date: Fri, 18 Jul 2008 08:41:22 To: Subject: [Comp-neuro] Review announcement Review announcement This review describes a constructive role for noise in synchronizing populations of neurons and should be of interest to computaional neurosciuentists. Trends Neurosci. 2008 Jul 4. [Epub ahead of print] Reliability, synchrony and noise. Ermentrout GB, Gal?n RF, Urban NN. The brain is noisy. Neurons receive tens of thousands of highly fluctuating inputs and generate spike trains that appear highly irregular. Much of this activity is spontaneous - uncoupled to overt stimuli or motor outputs - leading to questions about the functional impact of this noise. Although noise is most often thought of as disrupting patterned activity and interfering with the encoding of stimuli, recent theoretical and experimental work has shown that noise can play a constructive role - leading to increased reliability or regularity of neuronal firing in single neurons and across populations. These results raise fundamental questions about how noise can influence neural function and computation. PMID: 18603311 [PubMed - as supplied by publisher] http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_user=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_version=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From bower at uthscsa.edu Tue Jul 22 11:50:37 2008 From: bower at uthscsa.edu (jim bower) Date: Tue Jul 22 11:56:04 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <002701c8ebdd$eb4b9cc0$81c9f889@braun03> References: <20080718111439.4E18A8FEE3C@neuroinf.org><48808F72.80501@cmu.edu> <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry><002701c8ebdd$eb4b9cc0$81c9f889@braun03> Message-ID: <1040446318-1216720299-cardhu_decombobulator_blackberry.rim.net-1093665069-@bxe111.bisx.prod.on.blackberry> And I have a fairly simple question in return. Does anyone know of any type of engine, computational or other that is as thermodynaically efficient as the nervous system? Burns glucose, has 10 to the 12th neuronal components and doesn't generate enough heat to keep itself warm. Given that level of efficiency, why wouldn't you expect optimality. Second question, is there any case where it has or can be measured that the nervous system doesn't opperate at or very near physical limits, single photon detection, quantal limits in electrorecptors, just over brownian noise in the auditory system, etc. Why wouldn't one expect similar levels of performance computationally. Next, if spike coding in the nervous system were at optimal efficiencies, what would you predict? Signal indistinguishable from noise. Finally, there is no taskmaster as rigorous or demanding as selection. We tend not to think so - but this is mostly wishful thinking. The history of neuroscience and computational neuroscience is full of examples where "smart" practitioners have declared one or another aspect of the nervous system to be less than optimized only to find out that they simply weren't asking the right question or didn't understand themselves the circumstances or real computational demand. Jim Bower Sent via BlackBerry by AT&T -----Original Message----- From: "Hans A. Braun" Date: Tue, 22 Jul 2008 11:32:54 To: ; Nathan Urban; ; Subject: Re: [Comp-neuro] Review announcement Hi Jim, nice to hear from you with an interesting question. Here is a question back: Who says that the biological coding scheme is optimised in a way as engineers would do? I have been educated as an engineer. Thereby, I specifically have learnt how handle, if it cannot be avoided, such detrimental system properties like noise, nonlinearities and time delays because these can lead to unpredictable system behavior, including undesired oscillation and chaos ? what regularly can be seen in all kind of biological systems. If something similar would happen in a car or an airplane, the responsible engineer, deservedly, would immediately be fired. Could it be that the engineer in evolution has used a principally different strategy? What was/is his/her goal? Who knows or who is interested to find the answer? - or a more appropriate question ;-) ? Coming back to the original "noise" question: During all the years as experimental physiologist I have got hundreds of hours recordings of impulse sequences from different neurons ? and all look more or less noisy - whatever it means. Best wishes Hans Braun PS: if you are interested, here are two references to our work (an actual and an earlier paper): Finke C, Vollmer J, Postnova S, Braun HA (2008) Propagation effects of current and conductance noise in a model neuron with subthreshold oscillations. Mathematical Biosciences doi:10.1016/j.mbs.2008.03.007 Braun HA, Wissing H, Sch?fer K, Hirsch MC (1994). Oscillation and noise determine signal transduction in shark multimodal sensory cells. Nature 367: 270-273. The first one is a mathematical/computational approach which has very recently been published, so far only as online version. The second reference is to a much earlier experimental paper which demonstrates how the evolutionary engineer might have used oscillations and noise to achieve a particular sensitivity. This strategy, for whatever reasons, was only realized for sensory encoding in some evolutionary very old animals like sharks. Dr. Hans A. Braun, Institute of Physiology, Deutschhausstr. 2, D-35037 Marburg, Germany. Tel: +49 (0)6421-286 23 05, FAX: +49 (0)6421-286 6967, E-mail: braun@staff.uni-marburg.de URL: http://www.uni-marburg.de/physiology/braun and http://www.clabs.de see also: http://www.BioSim-Network.net ----- Original Message ----- From: "jim bower" To: "Nathan Urban" ; ; Sent: Friday, July 18, 2008 3:56 PM Subject: Re: [Comp-neuro] Review announcement > Haven't done this in a long time. But who says neurons are noisy? > > From the point of view of information theory, why isn't the apperance of noise expected in a highly optimized coding scheme? And why isn't synchrony to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > > Just curious. > > Jim bower > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Nathan Urban > > Date: Fri, 18 Jul 2008 08:41:22 > To: > Subject: [Comp-neuro] Review announcement > > > Review announcement > > This review describes a constructive role for noise in synchronizing > populations of neurons and should be of interest to computaional > neurosciuentists. > > > Trends Neurosci. 2008 Jul 4. [Epub ahead of print] > Reliability, synchrony and noise. > Ermentrout GB, Gal?n RF, Urban NN. > > The brain is noisy. Neurons receive tens of thousands of highly > fluctuating inputs and generate spike trains that appear highly > irregular. Much of this activity is spontaneous - uncoupled to overt > stimuli or motor outputs - leading to questions about the functional > impact of this noise. Although noise is most often thought of as > disrupting patterned activity and interfering with the encoding of > stimuli, recent theoretical and experimental work has shown that noise > can play a constructive role - leading to increased reliability or > regularity of neuronal firing in single neurons and across populations. > These results raise fundamental questions about how noise can influence > neural function and computation. > > PMID: 18603311 [PubMed - as supplied by publisher] > > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_us er=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_versio n=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > ---------------------------------------------------------------------------- ---- > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > From cl243 at cornell.edu Fri Jul 18 18:19:32 2008 From: cl243 at cornell.edu (Christiane Linster) Date: Tue Jul 22 12:00:40 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry> References: <20080718111439.4E18A8FEE3C@neuroinf.org><48808F72.80501@cmu.edu> <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry> Message-ID: <4880C294.1090106@cornell.edu> I will add my two cents to Jim's. Theoretical work showing that noise can be useful is not new, but rather old - my knowledge of it goes back at least 30 years when simulated annealing etc was much talked about. Noise helping synchronization is also not a novel idea. jim bower wrote: > Haven't done this in a long time. But who says neurons are noisy? > > From the point of view of information theory, why isn't the apperance of noise expected in a highly optimized coding scheme? And why isn't synchrony to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > > Just curious. > > Jim bower > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Nathan Urban > > Date: Fri, 18 Jul 2008 08:41:22 > To: > Subject: [Comp-neuro] Review announcement > > > Review announcement > > This review describes a constructive role for noise in synchronizing > populations of neurons and should be of interest to computaional > neurosciuentists. > > > Trends Neurosci. 2008 Jul 4. [Epub ahead of print] > Reliability, synchrony and noise. > Ermentrout GB, Gal?n RF, Urban NN. > > The brain is noisy. Neurons receive tens of thousands of highly > fluctuating inputs and generate spike trains that appear highly > irregular. Much of this activity is spontaneous - uncoupled to overt > stimuli or motor outputs - leading to questions about the functional > impact of this noise. Although noise is most often thought of as > disrupting patterned activity and interfering with the encoding of > stimuli, recent theoretical and experimental work has shown that noise > can play a constructive role - leading to increased reliability or > regularity of neuronal firing in single neurons and across populations. > These results raise fundamental questions about how noise can influence > neural function and computation. > > PMID: 18603311 [PubMed - as supplied by publisher] > > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_user=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_version=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > ------------------------------------------------------------------------ > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro -- ********************************* Christiane Linster Associate Professor Neurobiology and Behavior Cornell University Mudd Hall W245 607 2544331 Ithaca, NY 14853 cl243@cornell.edu www.cpl.cornell.edu From bcseet at ieee.org Sun Jul 20 06:20:21 2008 From: bcseet at ieee.org (Boon-Chong Seet) Date: Tue Jul 22 12:03:43 2008 Subject: [Comp-neuro] Final CFP: Sensor Networks and Ambient Intelligence (July 21, 2008) Message-ID: <008c01c8ea1f$ed9a0940$714d18d2@yourbbc104cd11> -------------------------------------------------------------------------------- CFP: 1st International Workshop on Sensor Networks and Ambient Intelligence -------------------------------------------------------------------------------- SeNAmI 2008 1st International Workshop on Sensor Networks and Ambient Intelligence In conjuction with PDCAT'08 http://www.cs.otago.ac.nz/pdcat08 December 1-4, 2008, Dunedin, New Zealand Call for Papers Sensor networks is an enabling technology of Ambient Intelligence. The pervasive nature of unobtrusive sensors distributed in the environment, either embedded or transportable by mobile carriers, enables fine-grain capture of environmental or ambient information that provides the basis of intelligence for higher-order cognitive systems, i.e. systems with capabilities to perceive, reason, learn, and react intelligently to their environment. Such systems in turn are envisioned to have wide ranging applications from intelligent wildlife and building structure monitoring, to humanistic and social endevours such as health and elderly care service provisioning. This workshop aims to bring together researchers from academia and industry to discuss recent research and technology advances in related areas, and from such engagement to foster or stimulate innovations in cross-disciplinary designs and methodologies in the fields of both sensor networks and ambient intelligence. Topics of interest include but are not limited to: - Cognitive wireless sensor networks - Cooperative and distributed sensor localization - Context-aware reasoning and inference for sensor/RFID-based systems - Intelligence support for sensor network management - Sensor networking in heterogeneous wireless environments - Sensor data fusion for ubiquitous embedded computing - Ambient intelligence system architectures, applications, and services - Testbed implementation and experimental trials Manuscript submission Papers reporting original and unpublished research results and experience are solicited. All paper submissions will be handled electronically via EasyChair. Please follow the IEEE Computer Society Press Proceedings Author Guidelines to prepare your papers. Maximum page length of accepted papers will be limited to 6 pages with main body text printed on 10 point font. All accepted papers will be included in conference proceedings of PDCAT'08, which will be published by IEEE Computer Society Press and automatically included in the IEEE Xplore digital library. The proceedings will also be cited by Engineering Information (EI). Selected best papers would be considered for publication in a special issue of the International Journal of Autonomous and Adaptive Communication Systems (IJAACS). Important dates Paper submission due : July 21, 2008 (extended) Acceptance notification : August 25, 2008 Camera-ready due : September 1, 2008 Workshop date : TBA For further details, please visit: http://senami08.aut.ac.nz -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080720/efc960d5/attachment.html From bressler at fau.edu Mon Jul 21 04:59:02 2008 From: bressler at fau.edu (Steven Bressler) Date: Tue Jul 22 12:04:30 2008 Subject: [Comp-neuro] PhD position investigating connectivity and dynamics of large-scale brain networks Message-ID: <00f601c8eadd$bb0f1920$331e5b83@opal> PHD POSITION IN COMPLEX SYSTEMS & BRAIN SCIENCES CENTER FOR COMPLEX SYSTEMS & BRAIN SCIENCES FLORIDA ATLANTIC UNIVERSITY Applications are currently being accepted for interdisciplinary PhD training under the joint mentorship of Drs. Steven Bressler (http://www.ccs.fau.edu/~bressler) and Viktor Jirsa (http://www.ccs.fau.edu/~jirsa). The program of training will take a theoretical approach to the study of neural and cognitive function. The position will focus on understanding the connectivity and dynamics of large-scale brain networks. It will emphasize: * the construction of artificial brain networks based on known anatomical connectivity * the generation of simulated network time-series data * functional connectivity and causality analysis of simulated time-series data in relation to the analysis of LFP, MEG, and fMRI time-series data The program offers competitive, multi-year stipends and tuition remission. Long-term research and training visits (up to six months) in partner institutions in France are optional and encouraged. The ideal candidate should have the following qualifications: * Quantitative training in physics, mathematics, computer science, or related fields * Programming experience (Matlab, C/C++, .) * English speaking and writing skills Previous experience in neuroscience is not required but would be an advantage. Interested students are encouraged to submit a CV, contact details of two referees, and a short statement of research interests to either Dr. Bressler (bressler@ccs.fau.edu) or Dr. Jirsa (jirsa@ccs.fau.edu). Please visit the Center for Complex Systems & Brain Sciences web site at http://www.ccs.fau.edu/. Committed to Equal Opportunities. The Center for Complex Systems & Brain Sciences at Florida Atlantic University is in Boca Raton, situated between West Palm Beach and Fort Lauderdale, with easy access to the beautiful beaches and rich cultural life of the Miami-Dade metropolitan area. From hitzler at aifb.uni-karlsruhe.de Mon Jul 21 16:56:03 2008 From: hitzler at aifb.uni-karlsruhe.de (Pascal Hitzler) Date: Tue Jul 22 12:20:09 2008 Subject: [Comp-neuro] CfP Journal Special Issue on Recurrent Neural Networks *DEADLINE EXTENDED* Message-ID: <4884A383.6030501@aifb.uni-karlsruhe.de> *DEADLINE EXTENDED DUE TO MULTIPLE REQUESTS* Final Call for Papers: Journal Special Issue on == Perspectives and Challenges for Recurrent Neural Networks == Guest Editors: Marco Gori, Barbara Hammer, Pascal Hitzler, Guenther Palm Special issue of the Elsevier Journal of Algorithms in Cognition, Informatics and Logic http://www.elsevier.com/wps/find/journaldescription.cws_home/622851/description = SCOPE = Recurrent neural networks (RNNs) enable flexible machine learning tools which can directly process spatiotemporal and other structured data and which offer a rich dynamic repertoire as time dependent systems. They promise to be efficient signal-processing models which are biologically plausible and optimally suited for a wide range of industrial applications on the one hand, and an explanation of cognitive phenomena of the human brain on the other hand. Despite these facts, however, the design of efficient training methods for RNNs as well as their mathematical investigation with respect to reliable information representation and generalization abilities when dealing with complex data structures is still a challenge. It has led to diverse approaches and architectures including echo and liquid-state-machines, long short term memory, recursive and graph networks, core neuro-symbolic integration, etc. Interestingly, very heterogeneous domains are included, such as logic, chaotic systems, and biological networks. The aim of the special issue is to bring together recent work developed in the field of recurrent information processing, which bridges the gap between different approaches and which sheds some light on canonical solutions or principled problems which occur in the context of recursive information processing when considered across the disciplines. = TOPICS = We particularly encourage submissions connected to the following non-exhaustive list of topics: - new learning paradigms of RNNs such as unsupervised learning or reservoire learning - biologically plausible methods - integration of RNNs and symbolic reasoning - universal approaches for general data structures such as sets or graphs - methods which address the generalization ability of RNNs - challenging applications which have the potential to be benchmark problems - visionary papers concerning the future of RNNs = SUBMISSIONS = Deadline for submissions is 22nd of August, 2008. Submissions shall follow the guidelines laid out for the Journal of Algorithms in Cognition, Informatics and Logic, which can be found under . Submissions shall be sent as pdf to Pascal Hitzler, hitzler@aifb.uni-karlsruhe.de = EDITORIAL BOARD = Guilherme da Alencar Barreto, Universidade Federal do Ceara, Brasil Monica Bianchini, University of Siena, Italy Howard Blair, Syracuse University, USA Hendrik Blockeel, KU Leuven, Belgium Mikael Boden, University of Queensland, Australia Matthew Cook, ETH Zuerich, Switzerland Artur d'Avila Garcez, City University London, UK Luc de Raedt, KU Leuven, Belgium Steffen Hoelldobler, TU Dresden, Germany Herbert Jaeger, Jacobs University Bremen, Germany Stefan C. Kremer, University of Guleph, Canada Kai-Uwe Kuehnberger, University of Osnabrueck, Germany Alessio Micheli, University of Pisa, Italy Barak Pearlmutter, NUI Maynooth, Ireland Juergen Schmidhuber, TU Munich, Germany Alessandro Sperduti, University of Padova, Italy Jochen Steil, University of Bielefeld, Germany Peter Tino, University of Bermingham, UK Edmondo Trentin, University of Siena, Italy Thomas Wennekers, University of Plymouth, UK This Call for Papers is available online under http://www.neural-symbolic.org/RNN_CfP.txt -- PD Dr. Pascal Hitzler Institute AIFB, University of Karlsruhe, 76128 Karlsruhe email: hitzler@aifb.uni-karlsruhe.de fax: +49 721 608 6580 web: http://www.pascal-hitzler.de phone: +49 721 608 4751 http://www.neural-symbolic.org From comar2 at illinois.edu Fri Jul 18 18:56:56 2008 From: comar2 at illinois.edu (Cyrus Omar) Date: Tue Jul 22 13:03:59 2008 Subject: [Comp-neuro] Key papers in computational neuroscience In-Reply-To: References: Message-ID: Here is the email without all the extra spaces: This is a collection of references obtained in response to a request for key papers from the computational neuroscience community. I have excluded self-citations (but many of those excluded papers actually appear in my own list of key papers below). I have removed the names of respondents, but have left their comments in, as these can be very useful. Many thanks to all those who contributed to this wide-ranging collection. Jim Stone, 18th July 2008. -- JV Stone's key papers: SB Laughlin. A simple coding procedure enhances a neuron's informationcapacity. Z Naturforsch, 36c:910{912, 1981.See other papers by Laughlin which cover similar material. Lettvin, J.Y., Maturana, H.R., McCulloch, W.S., and Pitts, W.H., What the Frog?s Eye Tells the Frog's Brain, Proc. Inst. Radio Engr. 47:1940-1951, 1959. Ballard, DH, Cortical connections and parallel processing: Structure and function, in Vision, in Brain and cooperative computation, pp 563-621, 1987, Arbib, MA and Hanson AR (Eds). Y Weiss, EP Simoncelli, and EH Adelson. Motion illusions as optimal percepts. Nature Neuroscience, 5(6):598 604, 2002. BA Olshausen and DJ Field. Sparse coding of sensory inputs. Current Opinion in Neurobiology, 14:481 487, 2004. T Poggio, V Torre, and C Koch. Computational vision and regularization theory. Nature, 317:314 319, 1985. AA Stocker and EP Simoncelli. Noise characteristics and prior expectations in human visual speed perception. Nature Neuroscience, 9(4):578 585, 2006. Marr, D., and T. Poggio. < http://cbcl.mit.edu/people/poggio/journals/marr-poggio-science-1976.pdf>Cooperative Computation of Stereo Disparity, Science, 194, 283-287, 1976. Rumelhart, D. E., Hinton, G. E., and Williams, R. J. Learning representations by back-propagating errors. Nature, 323, 533--536. Hinton, G. E. and Nowlan, S. J. How learning can guide evolution. Complex Systems, 1, 495--502. Hinton, G. E. and Plaut, D. C. Using fast weights to deblur old memories. Proceedings of the Ninth Annual Conference of the Cognitive Science Society, Seattle, WA Becker, S. and Hinton, G. E. A self-organizing neural network that discovers surfaces in random-dot stereograms. Nature, 355:6356, 161-163 Ackley, D. H., Hinton, G. E., and Sejnowski, T. J. A learning algorithm for Boltzmann machines. Cognitive Science, 9, 147-169. @article{DURBIN_WILLSHAW_TSP, author ="Durbin, R and Willshaw, D", title = "An analogue approach to the travelling salesman problem using an elastic net method", journal = "Nature", volume = "326", number = "6114", pages = "689-691", month = "", year = "1987" } @article{DOUGLAS_CANONICAL_89, author = "Douglas, RJ and Martin, KAC and Whitteridge, D", title = "A Canonical Microcircuit for Neocortex", journal = "Neural Computation", volume = "1", number = "", pages = "480-488", month = "", year = "1989" } @article{SWINDALE82, author = "Swindale, NV", title = "A model for the formation of orientation columns", journal = "Proceedings Royal Society London B", volume = "215", number = "", pages = "211-230", month = "", year = "1982" } Zohary, E, Shadlen, MN and Newsome, WT (1994). Correlated neuronal discharge rate and its implications for psychophysical performance. Nature 370:140-143. Hopfield's papers (see below). -- Hodgkin and Huxley 1952d (the modeling paper) -- Song and Abbott: Cortical development and remapping through spike timing-dependent plasticity.Neuron 32:339-50, 2001 and Buonomano and Merzevich: Temporal information transformed into a spatial code by a neural network with realistic properties.Science. 1995 Feb 17;267(5200):1028-30. -- Wilson HR, Cowan JD.Excitatory and inhibitory interactions in localized populations of modelneurons. Biophys J. 1972 Jan;12(1):1-24. H.B. Barlow, The mechanical mind.Ann. Rev. Neurosci. 13 15-24 (1990)It is about a simple model of consciousness. -- >From the cognitive side of computational neuroscience and I recommend: Pouget A, Deneve S, Duhamel JR (2002) A computational perspective on the neural basis of multisensory spatial representations. Nat Rev Neurosci. 3: 741-747. Hamker, F.H., Zirnsak, M., Calow, D., Lappe, M. (2008)?The peri-saccadic perception of objects and space.?PLOS Computational Biology 4(2):e21 Olshausen BA, Field DJ. 1996. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381:607-9. -- I was really influenced by @article{Atick92, Author = {Atick, Joseph J.}, Journal = {Network: {C}omputation in {N}eural {S}ystems}, Number = {2}, Pages = {213--52}, Title = {Could {I}nformation {T}heory {P}rovide an {E}cological {T}heory of {S}ensory {P}rocessing?}, Volume = {3}, Year = {1992}} which is a review paper rather related to the seminal papers from Barlow and Marr. -- Wilson HR, Cowan JD.Excitatory and inhibitory interactions in localized populations of modelneurons.Biophys J. 1972 Jan;12(1):1-24. -- Wiring Optimization Dmitri B. ChklovskiiTraub's CA1 model/Pinsky Rinzel 2 compartmental modelsErik De Schutter's Purkinje cell modelsHenry Markram's Cortical ModelsRolls & Treves - Hippocampal NetworkPolsky & Mel - 2layer pyramidal cell modelTerry Sejnowski - Synapse, modeldbUpinder S Bhalla - Million Synapses / Bistable systems -- These papers introduced accurate models of calcium dynamics and neuromodulatory effects on ion channel activity. Bhalla US, Iyengar R.Emergent properties of networks of biological signaling pathways.Science. 1999 Jan 15;283(5400):381-7. Zador A, Koch C, Brown TH.Biophysical model of a Hebbian synapse.Proc Natl Acad Sci U S A. 1990 Sep;87(17):6718-22. Holmes WR, Levy WB.AbstractInsights into associative long-term potentiation from computational models of NMDA receptor-mediated calcium influx and intracellular calcium concentration changes.J Neurophysiol. 1990 May;63(5):1148-68. -- There are two theoretical papers which, in my opinion, have had a strong influence on the way we think about synaptic transmission and short term plasticity today: A W Liley and K A North. An electrical investigation of effects of repetitivestimulation on mammalian neuromuscular junction. J Neurophysiol, 16(5):509 527, Sep 1953. W J Betz. Depression of transmitter release at the neuromuscular junction of thefrog. J Physiol, 206(3):629 644, 1970. These were, of course, published before the term "computation neuroscience" was used. The first proposed a mathematical model for vesicle pool depletion, which is still in use today. The second was the first to extend this with the release probability as a dynamic variable. These ideas were then further popularised by these classic papers: L F Abbott, J A Varela, K Sen, and S B Nelson. Synaptic depression and corticalgain control. Science, 275(5297):220 224, Jan 1997. M V Tsodyks and H Markram. The neural code between neocortical pyramidalneurons depends on neurotransmitter release probability. Proc Natl Acad Sci U SA, 94(2):719 723, Jan 1997. What I found have during my collaborations with biologists was that not so much the precise mathematical formulation, but the very basic ideas and concepts explored in these papers have made a strong impact in the whole field, and have certainly cleared the way for numerous further theoretical studies. Another paper I have come across just recently which I would consider as rather important and useful is this: J J Hopfield and A V M Herz. Rapid Local Synchronization of Action Potentials: Toward Computation with Coupled Integrate-and-Fire Neurons. Proc Natl Acad Sci U SA, 92(15): 6655-6662, Jul 1995. Cited more than 150 times, it contains some strong results regarding the behaviour of recurrent networks, and also anticipates a number of results shown more recently. -- Here is my top 12 papers, in chronological order. I have gone for ones that make my science heart sing, that introduce a big idea, useful tool, connect experiment and theory in a satisfying way, or are an example ofwork on a topic that has been mysteriously under-represented. I have tried to briefly qualify why they could be thought of as classic by the wider community. 1) Willshaw and von der Malsburg (1979). Future hot topic: modelling development Excellent interaction between theory and experiment - predicted ephrins and eph receptors. http://www.jstor.org/stable/pdfplus/2418226.pdf2) Laughlin (1981) Z. Naturforsch. C 36:910-2 Big idea: coding matches stimulus statistics. http://www.ncbi.nlm.nih.gov/sites/entrez?Db=pubmed&TermToSearch=73038233) Srinivasan et al. (1982) Proc. Roy. Soc. B 216(1205):427-59 Excellent interaction between theory and experiment: predicts responses of first order visual interneurons if they exploit spatial and temporal correlations to reduce redundancy. http://www.kyb.tuebingen.mpg.de/bethgegroup/teaching/ws0708_sem_retina_whitening/Srinivasan_et_al_1982.pdf 4) Buchsbaum and Gottschalk (1983). Proc. R. Soc. B 220:89-113 Excellent interaction between theory and experiment: uses PCA to accurately calculate the colour channels that maximise information transmission. Deserves to be more widely known. http://www.jstor.org/stable/pdfplus/35873.pdf 5) Bialek et al. (1991) Science Useful application for theorist: neat method for calculating stimulus filters in the response. http://www2.hawaii.edu/~sstill/neural_code_91.pdf 6) Treves and Rolls (1992) Hippocampus 2(2):189-99 Excellent interaction between theory and experiment: identified the function of the dentate gyrus in the hippocampus, and matched network organisation to function far more successfully that Marr. http://www3.interscience.wiley.com/cgi-bin/fulltext/109711333/PDFSTART7) Van Hateren (1992) J. Comp Phys. A 171:157-170 Excellent interaction between theory and experiment: predicts visual spatiotemporal receptive fields of cells connected to photoreceptors in the fly so as to maximise information about natural images from first principles, with stunning success. http://www.springerlink.com/content/h4681x344j378229/fulltext.pdf 8) Wolpert et al. (1995) Science 269(5232):1880-2 Big idea: internal models and the use of priors. http://keck.ucsf.edu/~houde/sensorimotor_jc/DMWolpert95a.pdf 9) Zemel et al. (1998) Neur. Comp. 10(2):403-30 Big idea: neurons encode distributions, not single values http://www.gatsby.ucl.ac.uk/~dayan/papers/zdp98.pdf 10) Van Rossum et al. (2000) J. Neuro. 20(23):8812-21 Excellent interaction between theory and experiment: Simple application of Fokker-Planck equation physics to explain functional consequences to the network of cellular level experimental data. http://www.jneurosci.org/cgi/reprint/20/23/8812.pdf 11) Brunel (2000) J. Comp. Neuro 8:183-208 Useful application for theorist: calculations of the population activity of a network of integrate-and-fire neurons. http://www.springerlink.com/content/u446l5722lp03677/fulltext.pdf 12) Schreiber (2000) Physical Review Letters 85(2):461-64 Future hot topic: Current best method to infer causal relationships between neurons using information theory. http://prola.aps.org/pdf/PRL/v85/i2/p461_1 -- Here are the most important papers in 3 subjects, plasticity andsimple neuron models and network dynamicsOf course, there are other categories in Computational neuroscience(detailed neuron model, cortex modeling, vision, audition etc) on which others will report. 1) In plasticity: Hebb, 1949 (book) Bienenstock, Cooper Munro, J. Neurosci.1982 (BCM rule) Kohonen Neural Networks1993 (Kohonen algo in comp neuro perspective other papers of him would also do)Hopfield, PNAS, 1982 (Hopfield model) Amit Gutfreund Sompolinksy, Phys Rev A, 1985 (Analysis of Hopfield model) Linsker PNAS, 1986 (emergence of field) MacKay and Miller 1990 Neural Comput. (analysis of Linskers rule) Miller and MacKay 1994 Neural Comput. the role of constraints Gerstner et al, Nature 1996 (first paper on STDP) Kempter et al. Phys Rev E, 1999 (first analysis of STDP) Lisman, PNAS, 1999 (first model of plasticity based on calcium dynamics) Song Miller Abbott, Nat. Neurosci, 2000 (popular paper on STDP) Rossum et al. 2000, J. Neuroscie (STDP with soft bounds for the weights) Fusi, Biological Cybernetics, 2002 (some general problems of Hebbian rules - nice review of work of Fusi)_ Shouval et al., PNAS, 2002 (calcium model of plasticity) Senn Tsodyks, Markram, Neural. comp. 2001 (STDP algorithm) Fusi, Drew, Abbott Nat. Neuroscience 2005 (Cascade model) Toyoizumi et al. PNAS 2005 (BCM rule for spiking neuron also optimized information) 2) In simplified neuron models Lapicque 07 (often cited as first integrate-and-fire model, even though it does not show reset) FitzHugh 1961, Biophys. Journal (2-dim neuron model) Stein 1967, Biophys. Journal (some models of neural variability - integrate-and-fire model with noise) Ermentrout 1996, Neural Comput., Canonical type I model, quadratic integrate-and-fire Kistler et al. 1997, Neural Computation (systematic reduction to a threshold model/Spike Response Model) Latham 2000, J. Neurophys. quadratic integrate-and-fire Izhikevich 2003, IEEE, 2-dim. neuron model Fourcaud et al. 2003, J. Neurosci. exp. integrate-and-fire model Jolivet et al. 2006, J. comput. Neurosci. -- spiking in real neurons can be explained by threshold models Badel et al. 2008, J. Neurophysiol. -- real neurons are exponential integrate-and-fire models, this is a very recent paper, but it is really important for the discussion of simple neuron models 3) Network dynamics Wilson and Cowan, 1972 Amari 1974 Brunel and Hakim, 1999 Neural Computation Gerstner 2000 Neural Computation Brunel 2000 Comput. Neurosci -- Finally, I am attaching a list of great papers. If I were trying to get outsiders excited, I'd definitely use the Andy Schwartz paper on neural prosthetics. Also think I would do Olshausen & Field as it really kicked people off on thinking about natural images. The Hopfield paper is the greatest of the bunch but is likely too old for what you're looking for. Spike-timing-dependent plasticity is a hot topic and I think carries on a great tradition of computational neuroscientists connecting cellular plasticity to larger network functions; and I think Peter Dayan (and Montague's in the original paper) work is some of the first that really puts a framework in place for thinking about neuromodulators. But they're all great, and I tried to hit many different contributions (maybe this is the greatest message--that computational neuroscience pervades so many fields from single-neuron computation to neuromodulators to models of memory). 1. Montague PR, Dayan P, Sejnowski TJ A framework for mesencephalic dopamine systems based on predictive Hebbian learning. J Neurosci. 1996 Mar 1;16(5):1936-47.Abstract: We develop a theoretical framework that shows how mesencephalic dopamine systems could distribute to their targets a signal that represents information about future expectations. In particular, we show how activity in the cerebral cortex can make predictions about future receipt of reward and how fluctuations in the activity levels of neurons in diffuse dopamine systems above and below baseline levels would represent errors in these predictions that are delivered to cortical and subcortical targets. We present a model for how such errors could be constructed in a real brain that is consistent with physiological results for a subset of dopaminergic neurons located in the ventral tegmental area and surrounding dopaminergic neurons. The theory also makes testable predictions about human choice behavior on a simple decision-making task. Furthermore, we show that, through a simple influence on synaptic plasticity, fluctuations in dopamine release can act to change the predictions in an appropriate manner.This paper is the first of a series of papers setting up a framework for how mesencephalic dopamine neurons represent reward and can serve as the basis for temporal difference -based reward learning in which the reward is offered at a delayed time. *2. Strong, S., Koberle, R., de Ruyter van Steveninck, R. and Bialek, W. 1998. Entropy and information in neural spike trains, Physical Review Letters 80: 197-200.Abstract. The nervous system represents time dependent signals in sequences of discrete, identical action potentials or spikes; information is carried only in the spike arrival times. We show how to quantify this information, in bits, free from any assumptions about which features of the spike train or input signal are most important, and we apply this approach to the analysis of experiments on a motion sensitive neuron in the fly visual system. This neuron transmits information about the visual stimulus at rates of up to 90 bits/s, within a factor of 2 of the physical limit set by the entropy of the spike train itself.This paper ushered in a new set of techniques for characterizing spike trains using the methods of information theory, and also illustrated that there was information on much smaller time scales (~a couple ms) than had typically been assumed previously. 3a. Abbott LF, Varela JA, Sen K, Nelson SB. Synaptic depression and cortical gain control.Science. 1997 Jan 10;275(5297):220-4Abstract. Cortical neurons receive synaptic inputs from thousands of afferents that fire action potentials at rates ranging from less than 1 hertz to more than 200 hertz. Both the number of afferents and their large dynamic range can mask changes in the spatial and temporal pattern of synaptic activity, limiting the ability of a cortical neuron to respond to its inputs. Modeling work based on experimental measurements indicates that short-term depression of intracortical synapses provides a dynamic gain-control mechanism that allows equal percentage rate changes on rapidly and slowly firing afferents to produce equal postsynaptic responses. Unlike inhibitory and adaptive mechanisms that reduce responsiveness to all inputs, synaptic depression is input-specific, leading to a dramatic increase in the sensitivity of a neuron to subtle changes in the firing patterns of its afferents. -AND- 3b. Markram H, Tsodyks M. Redistribution of synaptic efficacy between neocortical pyramidal neurons. Nature. 1996 Aug 29;382(6594):807-10.Abstract. Experience-dependent potentiation and depression of synaptic strength has been proposed to subserve learning and memory by changing the gain of signals conveyed between neurons. Here we examine synaptic plasticity between individual neocortical layer-5 pyramidal neurons. We show that an increase in the synaptic response, induced by pairing action-potential activity in pre- and postsynaptic neurons, was only observed when synaptic input occurred at low frequencies. This frequency-dependent increase in synaptic responses arises because of a redistribution of the available synaptic efficacy and not because of an increase in the efficacy. Redistribution of synaptic efficacy could represent a mechanism to change the content, rather than the gain, of signals conveyed between neurons.These 2 papers connected short-term synaptic plasticity to important computational implications. 4a. Hopfield JJ. Neural networks and physical systems with emergent collective computational abilities. Proc Natl Acad Sci U S A. 1982 Apr;79(8):2554-8Abstract. Computational properties of use of biological organisms or to the construction of computers can emerge as collective properties of systems having a large number of simple equivalent components (or neurons). The physical meaning of content-addressable memory is described by an appropriate phase space flow of the state of a system. A model of such a system is given, based on aspects of neurobiology but readily adapted to integrated circuits. The collective properties of this model produce a content-addressable memory which correctly yields an entire memory from any subpart of sufficient size. The algorithm for the time evolution of the state of the system is based on asynchronous parallel processing. Additional emergent collective properties include some capacity for generalization, familiarity recognition, categorization, error correction, and time sequence retention. The collective properties are only weakly sensitive to details of the modeling or the failure of individual devices.This classic paper illustrated the idea of attractor models and a correspondence with energy surfaces. It is now universally permeates discussions of long-term memory storage in networks, especially in the hippocampus. It was followed more recently by the article below, which expanded the idea of attractor models to continuous attractors this now is the framework for discussion of many networks storing short-term memories (the other set of models being the so-called ring models but i am not sure of the original reference for those). 4b. Seung HS. How the brain keeps the eyes still. Proc Natl Acad Sci U S A. 1996 Nov 12;93(23):13339-44.Abstract. The brain can hold the eyes still because it stores a memory of eye position. The brain's memory of horizontal eye position appears to be represented by persistent neural activity in a network known as the neural integrator, which is localized in the brainstem and cerebellum. Existing experimental data are reinterpreted as evidence for an "attractor hypothesis" that the persistent patterns of activity observed in this network form an attractive line of fixed points in its state space. Line attractor dynamics can be produced in linear or nonlinear neural networks by learning mechanisms that precisely tune positive feedback. 5a. Song S, Miller KD, Abbott LF. Competitive Hebbian learning through spike-timing-dependent synaptic plasticity. Nat Neurosci. 2000 Sep;3(9):919-26.Abstract. Hebbian models of development and learning require both activity-dependent synaptic plasticity and a mechanism that induces competition between different synapses. One form of experimentally observed long-term synaptic plasticity, which we call spike-timing-dependent plasticity (STDP), depends on the relative timing of pre- and postsynaptic action potentials. In modeling studies, we find that this form of synaptic modification can automatically balance synaptic strengths to make postsynaptic firing irregular but more sensitive to presynaptic spike timing. It has been argued that neurons in vivo operate in such a balanced regime. Synapses modifiable by STDP compete for control of the timing of postsynaptic action potentials. Inputs that fire the postsynaptic neuron with short latency or that act in correlated groups are able to compete most successfully and develop strong synapses, while synapses of longer-latency or less-effective inputs are weakened. -AND- 5b. Song S, Abbott LF.Neuron. Cortical development and remapping through spike timing-dependent plasticity. 2001 Oct 25;32(2):339-50Abstract. Long-term modification of synaptic efficacy can depend on the timing of pre- and postsynaptic action potentials. In model studies, such spike timing-dependent plasticity (STDP) introduces the desirable features of competition among synapses and regulation of postsynaptic firing characteristics. STDP strengthens synapses that receive correlated input, which can lead to the formation of stimulus-selective columns and the development, refinement, and maintenance of selectivity maps in network models. The temporal asymmetry of STDP suppresses strong destabilizing self-excitatory loops and allows a group of neurons that become selective early in development to direct other neurons to become similarly selective. STDP, acting alone without further hypothetical global constraints or additional forms of plasticity, can also reproduce the remapping seen in adult cortex following afferent lesions.The papers above have been seminal in illustrating the implications for learning of spike-timing-dependent synaptic plasticity 6. Polsky A, Mel BW, Schiller J. Nat Neurosci. 2004 Jun;7(6):621-7. Epub 2004 May 23.Computational subunits in thin dendrites of pyramidal cells.Abstract. The thin basal and oblique dendrites of cortical pyramidal neurons receive most of the synaptic inputs from other cells, but their integrative properties remain uncertain. Previous studies have most often reported global linear or sublinear summation. An alternative view, supported by biophysical modeling studies, holds that thin dendrites provide a layer of independent computational 'subunits' that sigmoidally modulate their inputs prior to global summation. To distinguish these possibilities, we combined confocal imaging and dual-site focal synaptic stimulation of identified thin dendrites in rat neocortical pyramidal neurons. We found that nearby inputs on the same branch summed sigmoidally, whereas widely separated inputs or inputs to different branches summed linearly. This strong spatial compartmentalization effect is incompatible with a global summation rule and provides the first experimental support for a two-layer 'neural network' model of pyramidal neuron thin-branch integration. Our findings could have important implications for the computing and memory-related functions of cortical tissue.This paper, as well as previous theoretical work, suggests that dendrites might enable single neurons to behave as feedforward neural networks. 7. Medina JF, Nores WL, Mauk MD. Nature. 2002 Mar 21;416(6878):330-3.Inhibition of climbing fibres is a signal for the extinction of conditioned eyelid responses.Abstract. A fundamental tenet of cerebellar learning theories asserts that climbing fibre afferents from the inferior olive provide a teaching signal that promotes the gradual adaptation of movements. Data from several forms of motor learning provide support for this tenet. In pavlovian eyelid conditioning, for example, where a tone is repeatedly paired with a reinforcing unconditioned stimulus like periorbital stimulation, the unconditioned stimulus promotes acquisition of conditioned eyelid responses by activating climbing fibres. Climbing fibre activity elicited by an unconditioned stimulus is inhibited during the expression of conditioned responses-consistent with the inhibitory projection from the cerebellum to inferior olive. Here, we show that inhibition of climbing fibres serves as a teaching signal for extinction, where learning not to respond is signalled by presenting a tone without the unconditioned stimulus. We used reversible infusion of synaptic receptor antagonists to show that blocking inhibitory input to the climbing fibres prevents extinction of the conditioned response, whereas blocking excitatory input induces extinction. These results, combined with analysis of climbing fibre activity in a computer simulation of the cerebellar-olivary system, suggest that transient inhibition of climbing fibres below their background level is the signal that drives extinction.This is one of several computational studies by Mauk and collaborators that are enhancing our knowledge of cerebellar processing (also see similar papers by Raymond & Lisberger applied to the VOR). 8. Olshausen BA, Field DJ. Nature. Emergence of simple-cell receptive field properties by learning a sparse code for natural images. 1996 Jun 13;381(6583):607-9Abstract.The receptive fields of simple cells in mammalian primary visual cortex can be characterized as being spatially localized, oriented and bandpass (selective to structure at different spatial scales), comparable to the basis functions of wavelet transforms. One approach to understanding such response properties of visual neurons has been to consider their relationship to the statistical structure of natural images in terms of efficient coding. Along these lines, a number of studies have attempted to train unsupervised learning algorithms on natural images in the hope of developing receptive fields with similar properties, but none has succeeded in producing a full set that spans the image space and contains all three of the above properties. Here we investigate the proposal that a coding strategy that maximizes sparseness is sufficient to account for these properties. We show that a learning algorithm that attempts to find sparse linear codes for natural scenes will develop a complete family of localized, oriented, bandpass receptive fields, similar to those found in the primary visual cortex. The resulting sparse image code provides a more efficient representation for later stages of processing because it possesses a higher degree of statistical independence among its outputs.This now classic study suggests how the statistical structure of natural images may determine the response properties of V1 cells, and set the stage for many later studies discussing the concept of sparse coding of images. 9. Taylor DM, Tillery SI, Schwartz AB. Direct cortical control of 3D neuroprosthetic devices. Science. 2002 Jun 7;296(5574):1829-32. Abstract. Three-dimensional (3D) movement of neuroprosthetic devices can be controlled by the activity of cortical neurons when appropriate algorithms are used to decode intended movement in real time. Previous studies assumed that neurons maintain fixed tuning properties, and the studies used subjects who were unaware of the movements predicted by their recorded units. In this study, subjects had real-time visual feedback of their brain-controlled trajectories. Cell tuning properties changed when used for brain-controlled movements. By using control algorithms that track these changes, subjects made long sequences of 3D movements using far fewer cortical units than expected. Daily practice improved movement accuracy and the directional tuning of these units.This represents some of the seminal work decoding cortical activity to control neural prosthetics. 10. Van Vreeswijk C, Abbott LF, Ermentrout GB. When inhibition not excitation synchronizes neural firing. J Comput Neurosci. 1994 Dec;1(4):313-21.Abstract. Excitatory and inhibitory synaptic coupling can have counter-intuitive effects on the synchronization of neuronal firing. While it might appear that excitatory coupling would lead to synchronization, we show that frequently inhibition rather than excitation synchronizes firing. We study two identical neurons described by integrate-and-fire models, general phase-coupled models or the Hodgkin-Huxley model with mutual, non-instantaneous excitatory or inhibitory synapses between them. We find that if the rise time of the synapse is longer than the duration of an action potential, inhibition not excitation leads to synchronized firing. If I were to update, I think I would add papers from: 1) Neuroeconomics & Reinforcement Learning-- in addition to the seminal work by Dayan & Schultz (already in attached), perhaps Loewenstein/Seung paper on matching behavior as a generic consequence of correlational learning rules. 2) Bayesian networks -- maybe Ma, Beck, Latham, Pouget or others on idea that the brain may encode & compute with probabilities --ENDOFFILE Cyrus On Fri, Jul 18, 2008 at 05:09, Jim Stone wrote: > > T h i s i s a c o l l e c t i o n o f r e f e r e n c e s o b t > a i n e d i n r e s p o n s e t o a r e q u e s t f o r k e y > p a p e r s f r o m t h e c o m p u t a t i o n a l n e u r o s c i e > n c e c o m m u n i t y . I h a v e e x c l u d e d s e l f - c i t > a t i o n s ( b u t m a n y o f t h o s e e x c l u d e d p a p e > r s a c t u a l l y a p p e a r i n m y o w n l i s t o f k e > y p a p e r s b e l o w ) . I h a v e r e m o v e d t h e n a m > e s o f r e s p o n d e n t s , b u t h a v e l e f t t h e i r > c o m m e n t s i n , a s t h e s e c a n b e v e r y u s e f u > l . > M a n y t h a n k s t o a l l t h o s e w h o c o n t r i b u t > e d t o t h i s w i d e - r a n g i n g c o l l e c t i o n . > J i m S t o n e , 1 8 t h J u l y 2 0 0 8 . > - - > J V S t o n e ' s k e y p a p e r s : > S B L a u g h l i n . A s i m p l e c o d i n g p r o c e d u r e > e n h a n c e s a n e u r o n ' s i n f o r m a t i o n c a p a c i t > y . Z N a t u r f o r s c h , 3 6 c : 9 1 0 { 9 1 2 , 1 9 8 1 . S e e > o t h e r p a p e r s b y L a u g h l i n w h i c h c o v e r s > i m i l a r m a t e r i a l . > L e t t v i n , J . Y . , M a t u r a n a , H . R . , M c C u l l o > c h , W . S . , a n d P i t t s , W . H . , W h a t t h e F r o > g ? s E y e T e l l s t h e F r o g ' s B r a i n , P r o c . I > n s t . R a d i o E n g r . 4 7 : 1 9 4 0 - 1 9 5 1 , 1 9 5 9 . > B a l l a r d , D H , C o r t i c a l c o n n e c t i o n s a n d > p a r a l l e l p r o c e s s i n g : S t r u c t u r e a n d f u n c > t i o n , i n V i s i o n , i n B r a i n a n d c o o p e r a t i > v e c o m p u t a t i o n , p p 5 6 3 - 6 2 1 , 1 9 8 7 , A r b i b > , M A a n d H a n s o n A R ( E d s ) . > Y W e i s s , E P S i m o n c e l l i , a n d E H A d e l s o n > . M o t i o n i l l u s i o n s a s o p t i m a l p e r c e p t s . > N a t u r e N e u r o s c i e n c e , 5 ( 6 ) : 5 9 8 6 0 4 , 2 0 0 > 2 . > B A O l s h a u s e n a n d D J F i e l d . S p a r s e c o d i > n g o f s e n s o r y i n p u t s . C u r r e n t O p i n i o n i > n N e u r o b i o l o g y , 1 4 : 4 8 1 4 8 7 , 2 0 0 4 . > T P o g g i o , V T o r r e , a n d C K o c h . C o m p u t a > t i o n a l v i s i o n a n d r e g u l a r i z a t i o n t h e o r y > . N a t u r e , 3 1 7 : 3 1 4 3 1 9 , 1 9 8 5 . > A A S t o c k e r a n d E P S i m o n c e l l i . N o i s e c h a > r a c t e r i s t i c s a n d p r i o r e x p e c t a t i o n s i n > h u m a n v i s u a l s p e e d p e r c e p t i o n . N a t u r e N > e u r o s c i e n c e , 9 ( 4 ) : 5 7 8 5 8 5 , 2 0 0 6 . > M a r r , D . , a n d T . P o g g i o . < h t t p : / / c b c l . > m i t . e d u / p e o p l e / p o g g i o / j o u r n a l s / m a r r - p o > g g i o - s c i e n c e - 1 9 7 6 . p d f > C o o p e r a t i v e C o m p > u t a t i o n o f S t e r e o D i s p a r i t y , S c i e n c e , 1 > 9 4 , 2 8 3 - 2 8 7 , 1 9 7 6 . > R u m e l h a r t , D . E . , H i n t o n , G . E . , a n d W i > l l i a m s , R . J . L e a r n i n g r e p r e s e n t a t i o n s > b y b a c k - p r o p a g a t i n g e r r o r s . N a t u r e , 3 2 3 > , 5 3 3 - - 5 3 6 . > H i n t o n , G . E . a n d N o w l a n , S . J . H o w l e a > r n i n g c a n g u i d e e v o l u t i o n . C o m p l e x S y s t > e m s , 1 , 4 9 5 - - 5 0 2 . > H i n t o n , G . E . a n d P l a u t , D . C . U s i n g > f a s t w e i g h t s t o d e b l u r o l d m e m o r i e s . P r > o c e e d i n g s o f t h e N i n t h A n n u a l C o n f e r e n c > e o f t h e C o g n i t i v e S c i e n c e S o c i e t y , S e a > t t l e , W A > B e c k e r , S . a n d H i n t o n , G . E . A s e l f - o > r g a n i z i n g n e u r a l n e t w o r k t h a t d i s c o v e r s > s u r f a c e s i n r a n d o m - d o t s t e r e o g r a m s . N a > t u r e , 3 5 5 : 6 3 5 6 , 1 6 1 - 1 6 3 > A c k l e y , D . H . , H i n t o n , G . E . , a n d S e j n > o w s k i , T . J . A l e a r n i n g a l g o r i t h m f o r > B o l t z m a n n m a c h i n e s . C o g n i t i v e S c i e n c e , > 9 , 1 4 7 - 1 6 9 . > @ a r t i c l e { D U R B I N _ W I L L S H A W _ T S P , a u t h o > r = " D u r b i n , R a n d W i l l s h a w , D " , t i t > l e = " A n a n a l o g u e a p p r o a c h t o t h e t r a > v e l l i n g s a l e s m a n p r o b l e m u s i n g a n e l a s t > i c n e t m e t h o d " , j o u r n a l = " N a t u r e " , > v o l u m e = " 3 2 6 " , n u m b e r = " 6 1 1 4 " > , p a g e s = " 6 8 9 - 6 9 1 " , m o n t h = > " " , y e a r = " 1 9 8 7 " } > @ a r t i c l e { D O U G L A S _ C A N O N I C A L _ 8 9 , a u t h > o r = " D o u g l a s , R J a n d M a r t i n , K A C a n d > W h i t t e r i d g e , D " , t i t l e = " A C a n o n i > c a l M i c r o c i r c u i t f o r N e o c o r t e x " , j o u > r n a l = " N e u r a l C o m p u t a t i o n " , v o l u m e > = " 1 " , n u m b e r = " " , p a g e s = " > 4 8 0 - 4 8 8 " , m o n t h = " " , y e a r = > " 1 9 8 9 " } > @ a r t i c l e { S W I N D A L E 8 2 , a u t h o r = " S w i > n d a l e , N V " , t i t l e = " A m o d e l f o r t > h e f o r m a t i o n o f o r i e n t a t i o n c o l u m n s " , > j o u r n a l = " P r o c e e d i n g s R o y a l S o c i e t y L > o n d o n B " , v o l u m e = " 2 1 5 " , n u m b e r > = " " , p a g e s = " 2 1 1 - 2 3 0 " , m o n t > h = " " , y e a r = " 1 9 8 2 " } > > Z o h a r y , E , S h a d l e n , M N a n d N e w s o m e , W T > ( 1 9 9 4 ) . C o r r e l a t e d n e u r o n a l d i s c h a r g e > r a t e a n d i t s i m p l i c a t i o n s f o r p s y c h o p h y > s i c a l p e r f o r m a n c e . N a t u r e 3 7 0 : 1 4 0 - 1 4 3 . > H o p f i e l d 's p a p e r s ( s e e b e l o w ) . > - - > H o d g k i n a n d H u x l e y 1 9 5 2 d ( t h e m o d e l i n g > p a p e r ) > - - > S o n g a n d A b b o t t : C o r t i c a l d e v e l o p m e n t > a n d r e m a p p i n g t h r o u g h s p i k e t i m i n g - d e p e > n d e n t p l a s t i c i t y . N e u r o n 3 2 : 3 3 9 - 5 0 , 2 0 0 1 > > a n d > B u o n o m a n o a n d M e r z e v i c h : T e m p o r a l i n f o > r m a t i o n t r a n s f o r m e d i n t o a s p a t i a l c o d e > b y a n e u r a l n e t w o r k w i t h r e a l i s t i c p r o > p e r t i e s . S c i e n c e . 1 9 9 5 F e b 1 7 ; 2 6 7 ( 5 2 0 0 ) > : 1 0 2 8 - 3 0 . > - - > W i l s o n H R , C o w a n J D . E x c i t a t o r y a n d i n h > i b i t o r y i n t e r a c t i o n s i n l o c a l i z e d p o p u l > a t i o n s o f m o d e l n e u r o n s . B i o p h y s J . 1 9 7 2 > J a n ; 1 2 ( 1 ) : 1 - 2 4 . > H . B . B a r l o w , T h e m e c h a n i c a l m i n d . A n n . > R e v . N e u r o s c i . 1 3 1 5 - 2 4 ( 1 9 9 0 ) I t i s a b > o u t a s i m p l e m o d e l o f c o n s c i o u s n e s s . > > - - > F r o m t h e c o g n i t i v e s i d e o f c o m p u t a t i o n > a l n e u r o s c i e n c e a n d I r e c o m m e n d : > P o u g e t A , D e n e v e S , D u h a m e l J R ( 2 0 0 2 ) > A c o m p u t a t i o n a l p e r s p e c t i v e o n t h e n e u r a > l b a s i s o f m u l t i s e n s o r y s p a t i a l r e p r e s e > n t a t i o n s . N a t R e v N e u r o s c i . 3 : 7 4 1 - 7 4 7 . > > H a m k e r , F . H . , Z i r n s a k , M . , C a l o w , D . , > L a p p e , M . ( 2 0 0 8 ) ? T h e p e r i - s a c c a d i c p e r c > e p t i o n o f o b j e c t s a n d s p a c e . ? P L O S C o m p u > t a t i o n a l B i o l o g y 4 ( 2 ) : e 2 1 > O l s h a u s e n B A , F i e l d D J . 1 9 9 6 . E m e r g e n c > e o f s i m p l e - c e l l r e c e p t i v e f i e l d p r o p e r > t i e s b y l e a r n i n g a s p a r s e c o d e f o r n a t u > r a l i m a g e s . N a t u r e 3 8 1 : 6 0 7 - 9 . > - - > I w a s r e a l l y i n f l u e n c e d b y > @ a r t i c l e { A t i c k 9 2 , A u t h o r = { A t i c k , > J o s e p h J . } , J o u r n a l = { N e t w o r k : { C } o > m p u t a t i o n i n { N } e u r a l { S } y s t e m s } , N u > m b e r = { 2 } , P a g e s = { 2 1 3 - - 5 2 } , T > i t l e = { C o u l d { I } n f o r m a t i o n { T } h e o r y { P > } r o v i d e a n { E } c o l o g i c a l { T } h e o r y o f { S } > e n s o r y { P } r o c e s s i n g ? } , V o l u m e = { 3 } , > Y e a r = { 1 9 9 2 } } > w h i c h i s a r e v i e w p a p e r r a t h e r r e l a t e d > t o t h e s e m i n a l p a p e r s f r o m B a r l o w a n d > M a r r . > - - > W i l s o n H R , C o w a n J D . E x c i t a t o r y a n d i n h > i b i t o r y i n t e r a c t i o n s i n l o c a l i z e d p o p u l > a t i o n s o f m o d e l n e u r o n s . B i o p h y s J . 1 9 7 2 > J a n ; 1 2 ( 1 ) : 1 - 2 4 . > - - > W i r i n g O p t i m i z a t i o n D m i t r i B . C h k l o v s k > i i T r a u b ' s C A 1 m o d e l / P i n s k y R i n z e l 2 c o m > p a r t m e n t a l m o d e l s E r i k D e S c h u t t e r ' s P u r > k i n j e c e l l m o d e l s H e n r y M a r k r a m ' s C o r t i > c a l M o d e l s R o l l s & T r e v e s - H i p p o c a m p a l > N e t w o r k P o l s k y & M e l - 2 l a y e r p y r a m i d a l > c e l l m o d e l T e r r y S e j n o w s k i - S y n a p s e , > m o d e l d b U p i n d e r S B h a l l a - M i l l i o n S y n a p > s e s / B i s t a b l e s y s t e m s > - - > T h e s e p a p e r s i n t r o d u c e d a c c u r a t e m o d e l > s o f c a l c i u m d y n a m i c s a n d n e u r o m o d u l a t o > r y e f f e c t s o n i o n c h a n n e l a c t i v i t y . > B h a l l a U S , I y e n g a r R . E m e r g e n t p r o p e r t i > e s o f n e t w o r k s o f b i o l o g i c a l s i g n a l i n g > p a t h w a y s . S c i e n c e . 1 9 9 9 J a n 1 5 ; 2 8 3 ( 5 4 0 0 > ) : 3 8 1 - 7 . > Z a d o r A , K o c h C , B r o w n T H . B i o p h y s i c a l > m o d e l o f a H e b b i a n s y n a p s e . P r o c N a t l A > c a d S c i U S A . 1 9 9 0 S e p ; 8 7 ( 1 7 ) : 6 7 1 8 - 2 2 . > H o l m e s W R , L e v y W B . A b s t r a c t I n s i g h t s i n > t o a s s o c i a t i v e l o n g - t e r m p o t e n t i a t i o n f > r o m c o m p u t a t i o n a l m o d e l s o f N M D A r e c e p t > o r - m e d i a t e d c a l c i u m i n f l u x a n d i n t r a c e l > l u l a r c a l c i u m c o n c e n t r a t i o n c h a n g e s . J N > e u r o p h y s i o l . 1 9 9 0 M a y ; 6 3 ( 5 ) : 1 1 4 8 - 6 8 . > - - > T h e r e a r e t w o t h e o r e t i c a l p a p e r s w h i c h , > i n m y o p i n i o n , h a v e h a d a s t r o n g i n f l u > e n c e o n t h e w a y w e t h i n k a b o u t s y n a p t i c > t r a n s m i s s i o n a n d s h o r t t e r m p l a s t i c i t y > t o d a y : > A W L i l e y a n d K A N o r t h . A n e l e c t r i c a l > i n v e s t i g a t i o n o f e f f e c t s o f r e p e t i t i v e > s t i m u l a t i o n o n m a m m a l i a n n e u r o m u s c u l a r > j u n c t i o n . J N e u r o p h y s i o l , 1 6 ( 5 ) : 5 0 9 5 2 7 > , S e p 1 9 5 3 . > W J B e t z . D e p r e s s i o n o f t r a n s m i t t e r r e > l e a s e a t t h e n e u r o m u s c u l a r j u n c t i o n o f > t h e f r o g . J P h y s i o l , 2 0 6 ( 3 ) : 6 2 9 6 4 4 , 1 9 > 7 0 . > T h e s e w e r e , o f c o u r s e , p u b l i s h e d b e f o r > e t h e t e r m " c o m p u t a t i o n n e u r o s c i e n c e " w > a s u s e d . T h e f i r s t p r o p o s e d a m a t h e m a t i > c a l m o d e l f o r v e s i c l e p o o l d e p l e t i o n , w > h i c h i s s t i l l i n u s e t o d a y . T h e s e c o n d > w a s t h e f i r s t t o e x t e n d t h i s w i t h t h e r > e l e a s e p r o b a b i l i t y a s a d y n a m i c v a r i a b l > e . T h e s e i d e a s w e r e t h e n f u r t h e r p o p u l a > r i s e d b y t h e s e c l a s s i c p a p e r s : > L F A b b o t t , J A V a r e l a , K S e n , a n d S B > N e l s o n . S y n a p t i c d e p r e s s i o n a n d c o r t i c > a l g a i n c o n t r o l . S c i e n c e , 2 7 5 ( 5 2 9 7 ) : 2 2 0 > 2 2 4 , J a n 1 9 9 7 . > M V T s o d y k s a n d H M a r k r a m . T h e n e u r a l > c o d e b e t w e e n n e o c o r t i c a l p y r a m i d a l n e u r o > n s d e p e n d s o n n e u r o t r a n s m i t t e r r e l e a s e > p r o b a b i l i t y . P r o c N a t l A c a d S c i U S A , > 9 4 ( 2 ) : 7 1 9 7 2 3 , J a n 1 9 9 7 . > W h a t I f o u n d h a v e d u r i n g m y c o l l a b o r a t > i o n s w i t h b i o l o g i s t s w a s t h a t n o t s o m u > c h t h e p r e c i s e m a t h e m a t i c a l f o r m u l a t i o n > , b u t t h e v e r y b a s i c i d e a s a n d c o n c e p t s > e x p l o r e d i n t h e s e p a p e r s h a v e m a d e a s > t r o n g i m p a c t i n t h e w h o l e f i e l d , a n d h a > v e c e r t a i n l y c l e a r e d t h e w a y f o r n u m e r o > u s f u r t h e r t h e o r e t i c a l s t u d i e s . > A n o t h e r p a p e r I h a v e c o m e a c r o s s j u s t r > e c e n t l y w h i c h I w o u l d c o n s i d e r a s r a t h e > r i m p o r t a n t a n d u s e f u l i s t h i s : > J J H o p f i e l d a n d A V M H e r z . R a p i d L o c > a l S y n c h r o n i z a t i o n o f A c t i o n P o t e n t i a l s > : T o w a r d C o m p u t a t i o n w i t h C o u p l e d I n t e g r > a t e - a n d - F i r e N e u r o n s . P r o c N a t l A c a d S c > i U S A , 9 2 ( 1 5 ) : 6 6 5 5 - 6 6 6 2 , J u l 1 9 9 5 . > C i t e d m o r e t h a n 1 5 0 t i m e s , i t c o n t a i n s > s o m e s t r o n g r e s u l t s r e g a r d i n g t h e b e h a > v i o u r o f r e c u r r e n t n e t w o r k s , a n d a l s o a > n t i c i p a t e s a n u m b e r o f r e s u l t s s h o w n m o > r e r e c e n t l y . > > - - > H e r e i s m y t o p 1 2 p a p e r s , i n c h r o n o l o g > i c a l o r d e r . I h a v e g o n e f o r o n e s t h a t m > a k e m y s c i e n c e h e a r t s i n g , t h a t i n t r o d u > c e a b i g i d e a , u s e f u l t o o l , c o n n e c t e x p > e r i m e n t a n d t h e o r y i n a s a t i s f y i n g w a y , > o r a r e a n e x a m p l e o f w o r k o n a t o p i c t h > a t h a s b e e n m y s t e r i o u s l y u n d e r - r e p r e s e n > t e d . > I h a v e t r i e d t o b r i e f l y q u a l i f y w h y t h > e y c o u l d b e t h o u g h t o f a s c l a s s i c b y t h > e w i d e r c o m m u n i t y . > 1 ) W i l l s h a w a n d v o n d e r M a l s b u r g ( 1 9 7 9 > ) . F u t u r e h o t t o p i c : m o d e l l i n g d e v e l > o p m e n t E x c e l l e n t i n t e r a c t i o n b e t w e e > n t h e o r y a n d e x p e r i m e n t - p r e d i c t e d > e p h r i n s a n d e p h r e c e p t o r s . h t t p : / / > w w w . j s t o r . o r g / s t a b l e / p d f p l u s / 2 4 1 8 2 2 6 . p > d f 2 ) L a u g h l i n ( 1 9 8 1 ) Z . N a t u r f o r s c h . C > 3 6 : 9 1 0 - 2 B i g i d e a : c o d i n g m a t c h e s s t > i m u l u s s t a t i s t i c s . h t t p : / / w w w . n c b i . n > l m . n i h . g o v / s i t e s / e n t r e z ? D b = p u b m e d & T e r m > T o S e a r c h = 7 3 0 3 8 2 3 3 ) S r i n i v a s a n e t a l . ( 1 > 9 8 2 ) P r o c . R o y . S o c . B 2 1 6 ( 1 2 0 5 ) : 4 2 7 - 5 9 > E x c e l l e n t i n t e r a c t i o n b e t w e e n t h e o r > y a n d e x p e r i m e n t : p r e d i c t s r e s p o n s e > s o f f i r s t o r d e r v i s u a l i n t e r n e u r o n s > i f t h e y e x p l o i t s p a t i a l a n d t e m p o r a l c > o r r e l a t i o n s t o r e d u c e r e d u n d a n c y . > h t t p : / / w w w . k y b . t u e b i n g e n . m p g . d e / b e t h > g e g r o u p / t e a c h i n g / w s 0 7 0 8 _ s e m _ r e t i n a _ w h i > t e n i n g / S r i n i v a s a n _ e t _ a l _ 1 9 8 2 . p d f > 4 ) B u c h s b a u m a n d G o t t s c h a l k ( 1 9 8 3 ) . P r > o c . R . S o c . B 2 2 0 : 8 9 - 1 1 3 E x c e l l e n t i > n t e r a c t i o n b e t w e e n t h e o r y a n d e x p e r i m e n > t : u s e s P C A t o a c c u r a t e l y c a l c u l a t e > t h e c o l o u r c h a n n e l s t h a t m a x i m i s e i n > f o r m a t i o n t r a n s m i s s i o n . D e s e r v e s t o b e > m o r e w i d e l y k n o w n . h t t p : / / w w w . > j s t o r . o r g / s t a b l e / p d f p l u s / 3 5 8 7 3 . p d f > > 5 ) B i a l e k e t a l . ( 1 9 9 1 ) S c i e n c e U s e > f u l a p p l i c a t i o n f o r t h e o r i s t : n e a t m e t h > o d f o r c a l c u l a t i n g s t i m u l u s f i l t e r s > i n t h e r e s p o n s e . h t t p : / / w w w 2 . h a w a i > i . e d u / ~ s s t i l l / n e u r a l _ c o d e _ 9 1 . p d f > > 6 ) T r e v e s a n d R o l l s ( 1 9 9 2 ) H i p p o c a m p u s > 2 ( 2 ) : 1 8 9 - 9 9 E x c e l l e n t i n t e r a c t i o n b > e t w e e n t h e o r y a n d e x p e r i m e n t : i d e n t i > f i e d t h e f u n c t i o n o f t h e d e n t a t e g y r u s > i n t h e h i p p o c a m p u s , a n d m a t c h e d n e t w > o r k o r g a n i s a t i o n t o f u n c t i o n f a r m o r > e s u c c e s s f u l l y t h a t M a r r . h t t p : / / w w w > 3 . i n t e r s c i e n c e . w i l e y . c o m / c g i - b i n / f u l l t > e x t / 1 0 9 7 1 1 3 3 3 / P D F S T A R T 7 ) V a n H a t e r e n ( > 1 9 9 2 ) J . C o m p P h y s . A 1 7 1 : 1 5 7 - 1 7 0 E x > c e l l e n t i n t e r a c t i o n b e t w e e n t h e o r y a n d > e x p e r i m e n t : p r e d i c t s v i s u a l s p a t i o t e > m p o r a l r e c e p t i v e f i e l d s o f c e l l s c o n > n e c t e d t o p h o t o r e c e p t o r s i n t h e f l y s o > a s t o m a x i m i s e i n f o r m a t i o n a b o u t n a > t u r a l i m a g e s f r o m f i r s t p r i n c i p l e s , > w i t h s t u n n i n g s u c c e s s . h t t p : / / w w w . > s p r i n g e r l i n k . c o m / c o n t e n t / h 4 6 8 1 x 3 4 4 j 3 7 8 > 2 2 9 / f u l l t e x t . p d f > > 8 ) W o l p e r t e t a l . ( 1 9 9 5 ) S c i e n c e 2 6 9 ( 5 > 2 3 2 ) : 1 8 8 0 - 2 B i g i d e a : i n t e r n a l m o d e l > s a n d t h e u s e o f p r i o r s . h t t p : / / k e c > k . u c s f . e d u / ~ h o u d e / s e n s o r i m o t o r _ j c / D M W o > l p e r t 9 5 a . p d f > > 9 ) Z e m e l e t a l . ( 1 9 9 8 ) N e u r . C o m p . 1 0 ( > 2 ) : 4 0 3 - 3 0 B i g i d e a : n e u r o n s e n c o d e d > i s t r i b u t i o n s , n o t s i n g l e v a l u e s h t t > p : / / w w w . g a t s b y . u c l . a c . u k / ~ d a y a n / p a p e r s > / z d p 9 8 . p d f > > 1 0 ) V a n R o s s u m e t a l . ( 2 0 0 0 ) J . N e u r o . > 2 0 ( 2 3 ) : 8 8 1 2 - 2 1 E x c e l l e n t i n t e r a c t i o > n b e t w e e n t h e o r y a n d e x p e r i m e n t : S i m > p l e a p p l i c a t i o n o f F o k k e r - P l a n c k e q u a t i > o n p h y s i c s t o e x p l a i n f u n c t i o n a l c o n > s e q u e n c e s t o t h e n e t w o r k o f c e l l u l a > r l e v e l e x p e r i m e n t a l d a t a . h t t p : / / > w w w . j n e u r o s c i . o r g / c g i / r e p r i n t / 2 0 / 2 3 / 8 8 > 1 2 . p d f > > 1 1 ) B r u n e l ( 2 0 0 0 ) J . C o m p . N e u r o 8 : 1 8 3 > - 2 0 8 U s e f u l a p p l i c a t i o n f o r t h e o r i s t > : c a l c u l a t i o n s o f t h e p o p u l a t i o n a c > t i v i t y o f a n e t w o r k o f i n t e g r a t e - a n d - f i > r e n e u r o n s . h t t p : / / w w w . s p r i n g e r l i n k > . c o m / c o n t e n t / u 4 4 6 l 5 7 2 2 l p 0 3 6 7 7 / f u l l t e x t > . p d f > > 1 2 ) S c h r e i b e r ( 2 0 0 0 ) P h y s i c a l R e v i e w L > e t t e r s 8 5 ( 2 ) : 4 6 1 - 6 4 F u t u r e h o t t o p i c > : C u r r e n t b e s t m e t h o d t o i n f e r c a u s a l > r e l a t i o n s h i p s b e t w e e n n e u r o n s u s i n g > i n f o r m a t i o n t h e o r y . h t t p : / / p r o l a . a > p s . o r g / p d f / P R L / v 8 5 / i 2 / p 4 6 1 _ 1 > - - > H e r e a r e t h e m o s t i m p o r t a n t p a p e r s i n > 3 s u b j e c t s , p l a s t i c i t y a n d s i m p l e n e u r o > n m o d e l s a n d n e t w o r k d y n a m i c s O f c o u r s e , > t h e r e a r e o t h e r c a t e g o r i e s i n C o m p u t a t > i o n a l n e u r o s c i e n c e ( d e t a i l e d n e u r o n m o d e > l , c o r t e x m o d e l i n g , v i s i o n , a u d i t i o n e t > c ) o n w h i c h o t h e r s w i l l r e p o r t . > 1 ) I n p l a s t i c i t y : > H e b b , 1 9 4 9 ( b o o k ) > B i e n e n s t o c k , C o o p e r M u n r o , J . N e u r o s c i > . 1 9 8 2 ( B C M r u l e ) > K o h o n e n N e u r a l N e t w o r k s 1 9 9 3 ( K o h o n e n a l > g o i n c o m p n e u r o p e r s p e c t i v e > o > t h e r p a p e r s o f h i m w o u l d a l s o d o ) H o p f i e > l d , P N A S , 1 9 8 2 ( H o p f i e l d m o d e l ) > A m i t G u t f r e u n d S o m p o l i n k s y , P h y s R e v A > , 1 9 8 5 ( A n a l y s i s o f H o p f i e l d m o d e l ) > L i n s k e r P N A S , 1 9 8 6 ( e m e r g e n c e o f f i e l d > ) > M a c K a y a n d M i l l e r 1 9 9 0 N e u r a l C o m p u t . > ( a n a l y s i s o f L i n s k e r s r u l e ) > M i l l e r a n d M a c K a y 1 9 9 4 N e u r a l C o m p u t . > t h e r o l e o f c o n s t r a i n t s > G e r s t n e r e t a l , N a t u r e 1 9 9 6 ( f i r s t p a p > e r o n S T D P ) > K e m p t e r e t a l . P h y s R e v E , 1 9 9 9 ( f i r s t > a n a l y s i s o f S T D P ) > L i s m a n , P N A S , 1 9 9 9 ( f i r s t m o d e l o f p l a s > t i c i t y b a s e d o n c a l c i u m d y n a m i c s ) > S o n g M i l l e r A b b o t t , N a t . N e u r o s c i , 2 0 0 > 0 ( p o p u l a r p a p e r o n S T D P ) > R o s s u m e t a l . 2 0 0 0 , J . N e u r o s c i e ( S T D P > w i t h s o f t b o u n d s f o r t h e w e i g h t s ) > F u s i , B i o l o g i c a l C y b e r n e t i c s , 2 0 0 2 ( s o > m e g e n e r a l p r o b l e m s o f H e b b i a n r u l e s - > n i c e r e v i e w o f w o r k o f F u s i ) _ > S h o u v a l e t a l . , P N A S , 2 0 0 2 ( c a l c i u m m o > d e l o f p l a s t i c i t y ) > S e n n T s o d y k s , M a r k r a m , N e u r a l . c o m p . 2 > 0 0 1 ( S T D P a l g o r i t h m ) > F u s i , D r e w , A b b o t t N a t . N e u r o s c i e n c e 2 > 0 0 5 ( C a s c a d e m o d e l ) > T o y o i z u m i e t a l . P N A S 2 0 0 5 ( B C M r u l e f > o r s p i k i n g n e u r o n a l s o o p t i m i z e d i n f o r m > a t i o n ) > > 2 ) I n s i m p l i f i e d n e u r o n m o d e l s > L a p i c q u e 0 7 ( o f t e n c i t e d a s f i r s t i n t e g > r a t e - a n d - f i r e m o d e l , e v e n t h o u g h i t d o e > s n o t s h o w r e s e t ) > F i t z H u g h 1 9 6 1 , B i o p h y s . J o u r n a l ( 2 - d i m > n e u r o n m o d e l ) > S t e i n 1 9 6 7 , B i o p h y s . J o u r n a l ( s o m e m o > d e l s o f n e u r a l v a r i a b i l i t y - i n t e g r a t e - > a n d - f i r e m o d e l w i t h n o i s e ) > E r m e n t r o u t 1 9 9 6 , N e u r a l C o m p u t . , C a n o n > i c a l t y p e I m o d e l , q u a d r a t i c i n t e g r a t e - > a n d - f i r e > K i s t l e r e t a l . 1 9 9 7 , N e u r a l C o m p u t a t i o > n ( s y s t e m a t i c r e d u c t i o n t o a t h r e s h o l d > m o d e l / S p i k e R e s p o n s e M o d e l ) > L a t h a m 2 0 0 0 , J . N e u r o p h y s . q u a d r a t i c i n > t e g r a t e - a n d - f i r e > I z h i k e v i c h 2 0 0 3 , I E E E , 2 - d i m . n e u r o n m > o d e l > F o u r c a u d e t a l . 2 0 0 3 , J . N e u r o s c i . e x p > . i n t e g r a t e - a n d - f i r e m o d e l > J o l i v e t e t a l . 2 0 0 6 , J . c o m p u t . N e u r o s > c i . - - s p i k i n g i n r e a l n e u r o n s c a n b e e > x p l a i n e d b y t h r e s h o l d m o d e l s > B a d e l e t a l . 2 0 0 8 , J . N e u r o p h y s i o l . - - > r e a l n e u r o n s a r e e x p o n e n t i a l i n t e g r a t e > - a n d - f i r e m o d e l s , t h i s i s a v e r y r e c e n t > p a p e r , > b u t i t i s r e a l > l y i m p o r t a n t f o r t h e d i s c u s s i o n o f s i m p > l e n e u r o n m o d e l s > > 3 ) N e t w o r k d y n a m i c s > W i l s o n a n d C o w a n , 1 9 7 2 > A m a r i 1 9 7 4 > B r u n e l a n d H a k i m , 1 9 9 9 N e u r a l C o m p u t a t > i o n > G e r s t n e r 2 0 0 0 N e u r a l C o m p u t a t i o n > B r u n e l 2 0 0 0 C o m p u t . N e u r o s c i > > - - > F i n a l l y , I a m a t t a c h i n g a l i s t o f g r e a > t p a p e r s . I f I w e r e t r y i n g t o g e t o u t s > i d e r s e x c i t e d , I ' d d e f i n i t e l y u s e t h e A > n d y S c h w a r t z p a p e r o n n e u r a l p r o s t h e t i c > s . A l s o t h i n k I w o u l d d o O l s h a u s e n & F > i e l d a s i t r e a l l y k i c k e d p e o p l e o f f o n > t h i n k i n g a b o u t n a t u r a l i m a g e s . T h e H o > p f i e l d p a p e r i s t h e g r e a t e s t o f t h e b u n > c h b u t i s l i k e l y t o o o l d f o r w h a t y o u ' r > e l o o k i n g f o r . S p i k e - t i m i n g - d e p e n d e n t p > l a s t i c i t y i s a h o t t o p i c a n d I t h i n k c a > r r i e s o n a g r e a t t r a d i t i o n o f c o m p u t a t i > o n a l n e u r o s c i e n t i s t s c o n n e c t i n g c e l l u l a > r p l a s t i c i t y t o l a r g e r n e t w o r k f u n c t i o n > s ; a n d I t h i n k P e t e r D a y a n ( a n d M o n t a g u > e ' s i n t h e o r i g i n a l p a p e r ) w o r k i s s o m e > o f t h e f i r s t t h a t r e a l l y p u t s a f r a m e w > o r k i n p l a c e f o r t h i n k i n g a b o u t n e u r o m o > d u l a t o r s . B u t t h e y ' r e a l l g r e a t , a n d I > t r i e d t o h i t m a n y d i f f e r e n t c o n t r i b u t i > o n s ( m a y b e t h i s i s t h e g r e a t e s t m e s s a g e > - - t h a t c o m p u t a t i o n a l n e u r o s c i e n c e p e r v a > d e s s o m a n y f i e l d s f r o m s i n g l e - n e u r o n c o > m p u t a t i o n t o n e u r o m o d u l a t o r s t o m o d e l s > o f m e m o r y ) . > 1 . M o n t a g u e P R , D a y a n P , S e j n o w s k i T J > A f r a m e w o r k f o r m e s e n c e p h a l i c d o p a m i n e > s y s t e m s b a s e d o n p r e d i c t i v e H e b b i a n l e a > r n i n g . J N e u r o s c i . 1 9 9 6 M a r 1 ; 1 6 ( 5 ) : 1 9 > 3 6 - 4 7 . A b s t r a c t : W e d e v e l o p a t h e o r e t i c a > l f r a m e w o r k t h a t s h o w s h o w m e s e n c e p h a l i > c d o p a m i n e s y s t e m s c o u l d d i s t r i b u t e t o > t h e i r t a r g e t s a s i g n a l t h a t r e p r e s e n t s > i n f o r m a t i o n a b o u t f u t u r e e x p e c t a t i o n s . > I n p a r t i c u l a r , w e s h o w h o w a c t i v i t y i n > t h e c e r e b r a l c o r t e x c a n m a k e p r e d i c t i o > n s a b o u t f u t u r e r e c e i p t o f r e w a r d a n d h > o w f l u c t u a t i o n s i n t h e a c t i v i t y l e v e l s > o f n e u r o n s i n d i f f u s e d o p a m i n e s y s t e m s > a b o v e a n d b e l o w b a s e l i n e l e v e l s w o u l d r > e p r e s e n t e r r o r s i n t h e s e p r e d i c t i o n s t h > a t a r e d e l i v e r e d t o c o r t i c a l a n d s u b c o r > t i c a l t a r g e t s . W e p r e s e n t a m o d e l f o r h > o w s u c h e r r o r s c o u l d b e c o n s t r u c t e d i n > a r e a l b r a i n t h a t i s c o n s i s t e n t w i t h p h > y s i o l o g i c a l r e s u l t s f o r a s u b s e t o f d o p > a m i n e r g i c n e u r o n s l o c a t e d i n t h e v e n t r a > l t e g m e n t a l a r e a a n d s u r r o u n d i n g d o p a m i > n e r g i c n e u r o n s . T h e t h e o r y a l s o m a k e s t > e s t a b l e p r e d i c t i o n s a b o u t h u m a n c h o i c e > b e h a v i o r o n a s i m p l e d e c i s i o n - m a k i n g t a > s k . F u r t h e r m o r e , w e s h o w t h a t , t h r o u g h > a s i m p l e i n f l u e n c e o n s y n a p t i c p l a s t i c i > t y , f l u c t u a t i o n s i n d o p a m i n e r e l e a s e c a > n a c t t o c h a n g e t h e p r e d i c t i o n s i n a n a > p p r o p r i a t e m a n n e r . T h i s p a p e r i s t h e f i r > s t o f a s e r i e s o f p a p e r s s e t t i n g u p a f > r a m e w o r k f o r h o w m e s e n c e p h a l i c d o p a m i n e > n e u r o n s r e p r e s e n t r e w a r d a n d c a n s e r v e > a s t h e b a s i s f o r t e m p o r a l d i f f e r e n c e - > b a s e d r e w a r d l e a r n i n g i n w h i c h t h e r e w a > r d i s o f f e r e d a t a d e l a y e d t i m e . > * 2 . S t r o n g , S . , K o b e r l e , R . , d e R u y t e > r v a n S t e v e n i n c k , R . a n d B i a l e k , W . 1 9 9 > 8 . E n t r o p y a n d i n f o r m a t i o n i n n e u r a l s p > i k e t r a i n s , P h y s i c a l R e v i e w L e t t e r s 8 0 : > 1 9 7 - 2 0 0 . A b s t r a c t . T h e n e r v o u s s y s t e m r > e p r e s e n t s t i m e d e p e n d e n t s i g n a l s i n s e q > u e n c e s o f d i s c r e t e , i d e n t i c a l a c t i o n p o > t e n t i a l s o r s p i k e s ; i n f o r m a t i o n i s c a r r > i e d o n l y i n t h e s p i k e a r r i v a l t i m e s . W e > s h o w h o w t o q u a n t i f y t h i s i n f o r m a t i o n , > i n b i t s , f r e e f r o m a n y a s s u m p t i o n s a b o > u t w h i c h f e a t u r e s o f t h e s p i k e t r a i n o r > i n p u t s i g n a l a r e m o s t i m p o r t a n t , a n d w > e a p p l y t h i s a p p r o a c h t o t h e a n a l y s i s o > f e x p e r i m e n t s o n a m o t i o n s e n s i t i v e n e u > r o n i n t h e f l y v i s u a l s y s t e m . T h i s n e u r > o n t r a n s m i t s i n f o r m a t i o n a b o u t t h e v i s u > a l s t i m u l u s a t r a t e s o f u p t o 9 0 b i t s / s > , w i t h i n a f a c t o r o f 2 o f t h e p h y s i c a l > l i m i t s e t b y t h e e n t r o p y o f t h e s p i k e t > r a i n i t s e l f . T h i s p a p e r u s h e r e d i n a n e w > s e t o f t e c h n i q u e s f o r c h a r a c t e r i z i n g s > p i k e t r a i n s u s i n g t h e m e t h o d s o f i n f o r m > a t i o n t h e o r y , a n d a l s o i l l u s t r a t e d t h a t > t h e r e w a s i n f o r m a t i o n o n m u c h s m a l l e r > t i m e s c a l e s ( ~ a c o u p l e m s ) t h a n h a d t y p > i c a l l y b e e n a s s u m e d p r e v i o u s l y . > 3 a . A b b o t t L F , V a r e l a J A , S e n K , N e l s o > n S B . S y n a p t i c d e p r e s s i o n a n d c o r t i c a l > g a i n c o n t r o l . S c i e n c e . 1 9 9 7 J a n 1 0 ; 2 7 5 ( 5 > 2 9 7 ) : 2 2 0 - 4 A b s t r a c t . C o r t i c a l n e u r o n s r e > c e i v e s y n a p t i c i n p u t s f r o m t h o u s a n d s o f > a f f e r e n t s t h a t f i r e a c t i o n p o t e n t i a l s > a t r a t e s r a n g i n g f r o m l e s s t h a n 1 h e r t z > t o m o r e t h a n 2 0 0 h e r t z . B o t h t h e n u m b e > r o f a f f e r e n t s a n d t h e i r l a r g e d y n a m i c > r a n g e c a n m a s k c h a n g e s i n t h e s p a t i a l > a n d t e m p o r a l p a t t e r n o f s y n a p t i c a c t i v i > t y , l i m i t i n g t h e a b i l i t y o f a c o r t i c a l > n e u r o n t o r e s p o n d t o i t s i n p u t s . M o d e l i > n g w o r k b a s e d o n e x p e r i m e n t a l m e a s u r e m e > n t s i n d i c a t e s t h a t s h o r t - t e r m d e p r e s s i o > n o f i n t r a c o r t i c a l s y n a p s e s p r o v i d e s a > d y n a m i c g a i n - c o n t r o l m e c h a n i s m t h a t a l l > o w s e q u a l p e r c e n t a g e r a t e c h a n g e s o n r a > p i d l y a n d s l o w l y f i r i n g a f f e r e n t s t o p r > o d u c e e q u a l p o s t s y n a p t i c r e s p o n s e s . U n l > i k e i n h i b i t o r y a n d a d a p t i v e m e c h a n i s m s > t h a t r e d u c e r e s p o n s i v e n e s s t o a l l i n p u t > s , s y n a p t i c d e p r e s s i o n i s i n p u t - s p e c i f i > c , l e a d i n g t o a d r a m a t i c i n c r e a s e i n t h > e s e n s i t i v i t y o f a n e u r o n t o s u b t l e c h a > n g e s i n t h e f i r i n g p a t t e r n s o f i t s a f f e > r e n t s . > > - A N D - > 3 b . M a r k r a m H , T s o d y k s M . R e d i s t r i b u t i > o n o f s y n a p t i c e f f i c a c y b e t w e e n n e o c o r t > i c a l p y r a m i d a l n e u r o n s . N a t u r e . 1 9 9 6 A u > g 2 9 ; 3 8 2 ( 6 5 9 4 ) : 8 0 7 - 1 0 . A b s t r a c t . E x p e r i e > n c e - d e p e n d e n t p o t e n t i a t i o n a n d d e p r e s s i > o n o f s y n a p t i c s t r e n g t h h a s b e e n p r o p o s > e d t o s u b s e r v e l e a r n i n g a n d m e m o r y b y c > h a n g i n g t h e g a i n o f s i g n a l s c o n v e y e d b e > t w e e n n e u r o n s . H e r e w e e x a m i n e s y n a p t i c > p l a s t i c i t y b e t w e e n i n d i v i d u a l n e o c o r t i > c a l l a y e r - 5 p y r a m i d a l n e u r o n s . W e s h o w > t h a t a n i n c r e a s e i n t h e s y n a p t i c r e s p o n > s e , i n d u c e d b y p a i r i n g a c t i o n - p o t e n t i a l > a c t i v i t y i n p r e - a n d p o s t s y n a p t i c n e u r > o n s , w a s o n l y o b s e r v e d w h e n s y n a p t i c i n > p u t o c c u r r e d a t l o w f r e q u e n c i e s . T h i s f > r e q u e n c y - d e p e n d e n t i n c r e a s e i n s y n a p t i c > r e s p o n s e s a r i s e s b e c a u s e o f a r e d i s t r i > b u t i o n o f t h e a v a i l a b l e s y n a p t i c e f f i c a > c y a n d n o t b e c a u s e o f a n i n c r e a s e i n t h > e e f f i c a c y . R e d i s t r i b u t i o n o f s y n a p t i c > e f f i c a c y c o u l d r e p r e s e n t a m e c h a n i s m t o > c h a n g e t h e c o n t e n t , r a t h e r t h a n t h e g a > i n , o f s i g n a l s c o n v e y e d b e t w e e n n e u r o n s > . T h e s e 2 p a p e r s c o n n e c t e d s h o r t - t e r m s y > n a p t i c p l a s t i c i t y t o i m p o r t a n t c o m p u t a t > i o n a l i m p l i c a t i o n s . > 4 a . H o p f i e l d J J . N e u r a l n e t w o r k s a n d p > h y s i c a l s y s t e m s w i t h e m e r g e n t c o l l e c t i v > e c o m p u t a t i o n a l a b i l i t i e s . P r o c N a t l A c > a d S c i U S A . 1 9 8 2 A p r ; 7 9 ( 8 ) : 2 5 5 4 - 8 A b s t > r a c t . C o m p u t a t i o n a l p r o p e r t i e s o f u s e o > f b i o l o g i c a l o r g a n i s m s o r t o t h e c o n s t r > u c t i o n o f c o m p u t e r s c a n e m e r g e a s c o l l e > c t i v e p r o p e r t i e s o f s y s t e m s h a v i n g a l a > r g e n u m b e r o f s i m p l e e q u i v a l e n t c o m p o n e > n t s ( o r n e u r o n s ) . T h e p h y s i c a l m e a n i n g > o f c o n t e n t - a d d r e s s a b l e m e m o r y i s d e s c r i > b e d b y a n a p p r o p r i a t e p h a s e s p a c e f l o w > o f t h e s t a t e o f a s y s t e m . A m o d e l o f s u > c h a s y s t e m i s g i v e n , b a s e d o n a s p e c t s > o f n e u r o b i o l o g y b u t r e a d i l y a d a p t e d t o > i n t e g r a t e d c i r c u i t s . T h e c o l l e c t i v e p r o > p e r t i e s o f t h i s m o d e l p r o d u c e a c o n t e n t > - a d d r e s s a b l e m e m o r y w h i c h c o r r e c t l y y i e > l d s a n e n t i r e m e m o r y f r o m a n y s u b p a r t o > f s u f f i c i e n t s i z e . T h e a l g o r i t h m f o r t h > e t i m e e v o l u t i o n o f t h e s t a t e o f t h e s y > s t e m i s b a s e d o n a s y n c h r o n o u s p a r a l l e l > p r o c e s s i n g . A d d i t i o n a l e m e r g e n t c o l l e c > t i v e p r o p e r t i e s i n c l u d e s o m e c a p a c i t y f > o r g e n e r a l i z a t i o n , f a m i l i a r i t y r e c o g n i t > i o n , c a t e g o r i z a t i o n , e r r o r c o r r e c t i o n , > a n d t i m e s e q u e n c e r e t e n t i o n . T h e c o l l e c > t i v e p r o p e r t i e s a r e o n l y w e a k l y s e n s i t i > v e t o d e t a i l s o f t h e m o d e l i n g o r t h e f a > i l u r e o f i n d i v i d u a l d e v i c e s . T h i s c l a s s i > c p a p e r i l l u s t r a t e d t h e i d e a o f a t t r a c t > o r m o d e l s a n d a c o r r e s p o n d e n c e w i t h e n e > r g y s u r f a c e s . I t i s n o w u n i v e r s a l l y p e > r m e a t e s d i s c u s s i o n s o f l o n g - t e r m m e m o r y > s t o r a g e i n n e t w o r k s , e s p e c i a l l y i n t h e > h i p p o c a m p u s . I t w a s f o l l o w e d m o r e r e c > e n t l y b y t h e a r t i c l e b e l o w , w h i c h e x p a n > d e d t h e i d e a o f a t t r a c t o r m o d e l s t o c o n > t i n u o u s a t t r a c t o r s t h i s n o w i s t h e f r a m > e w o r k f o r d i s c u s s i o n o f m a n y n e t w o r k s s > t o r i n g s h o r t - t e r m m e m o r i e s ( t h e o t h e r s > e t o f m o d e l s b e i n g t h e s o - c a l l e d r i n g m > o d e l s b u t i a m n o t s u r e o f t h e o r i g i n a > l r e f e r e n c e f o r t h o s e ) . > 4 b . S e u n g H S . H o w t h e b r a i n k e e p s t h e > e y e s s t i l l . P r o c N a t l A c a d S c i U S A . 1 > 9 9 6 N o v 1 2 ; 9 3 ( 2 3 ) : 1 3 3 3 9 - 4 4 . A b s t r a c t . T h > e b r a i n c a n h o l d t h e e y e s s t i l l b e c a u s e > i t s t o r e s a m e m o r y o f e y e p o s i t i o n . T h > e b r a i n ' s m e m o r y o f h o r i z o n t a l e y e p o s i > t i o n a p p e a r s t o b e r e p r e s e n t e d b y p e r s i s > t e n t n e u r a l a c t i v i t y i n a n e t w o r k k n o w n > a s t h e n e u r a l i n t e g r a t o r , w h i c h i s l o c > a l i z e d i n t h e b r a i n s t e m a n d c e r e b e l l u m . > E x i s t i n g e x p e r i m e n t a l d a t a a r e r e i n t e r > p r e t e d a s e v i d e n c e f o r a n " a t t r a c t o r h y > p o t h e s i s " t h a t t h e p e r s i s t e n t p a t t e r n s > o f a c t i v i t y o b s e r v e d i n t h i s n e t w o r k f o > r m a n a t t r a c t i v e l i n e o f f i x e d p o i n t s i > n i t s s t a t e s p a c e . L i n e a t t r a c t o r d y n a m > i c s c a n b e p r o d u c e d i n l i n e a r o r n o n l i n > e a r n e u r a l n e t w o r k s b y l e a r n i n g m e c h a n i > s m s t h a t p r e c i s e l y t u n e p o s i t i v e f e e d b a > c k . > 5 a . S o n g S , M i l l e r K D , A b b o t t L F . C o m p > e t i t i v e H e b b i a n l e a r n i n g t h r o u g h s p i k e - > t i m i n g - d e p e n d e n t s y n a p t i c p l a s t i c i t y . N > a t N e u r o s c i . 2 0 0 0 S e p ; 3 ( 9 ) : 9 1 9 - 2 6 . A b s t r > a c t . H e b b i a n m o d e l s o f d e v e l o p m e n t a n d > l e a r n i n g r e q u i r e b o t h a c t i v i t y - d e p e n d e > n t s y n a p t i c p l a s t i c i t y a n d a m e c h a n i s m > t h a t i n d u c e s c o m p e t i t i o n b e t w e e n d i f f e r > e n t s y n a p s e s . O n e f o r m o f e x p e r i m e n t a l l > y o b s e r v e d l o n g - t e r m s y n a p t i c p l a s t i c i t > y , w h i c h w e c a l l s p i k e - t i m i n g - d e p e n d e n t > p l a s t i c i t y ( S T D P ) , d e p e n d s o n t h e r e l a > t i v e t i m i n g o f p r e - a n d p o s t s y n a p t i c a c > t i o n p o t e n t i a l s . I n m o d e l i n g s t u d i e s , w > e f i n d t h a t t h i s f o r m o f s y n a p t i c m o d i f > i c a t i o n c a n a u t o m a t i c a l l y b a l a n c e s y n a p > t i c s t r e n g t h s t o m a k e p o s t s y n a p t i c f i r i > n g i r r e g u l a r b u t m o r e s e n s i t i v e t o p r e s > y n a p t i c s p i k e t i m i n g . I t h a s b e e n a r g u e > d t h a t n e u r o n s i n v i v o o p e r a t e i n s u c h > a b a l a n c e d r e g i m e . S y n a p s e s m o d i f i a b l e > b y S T D P c o m p e t e f o r c o n t r o l o f t h e t i m > i n g o f p o s t s y n a p t i c a c t i o n p o t e n t i a l s . > I n p u t s t h a t f i r e t h e p o s t s y n a p t i c n e u r o > n w i t h s h o r t l a t e n c y o r t h a t a c t i n c o r > r e l a t e d g r o u p s a r e a b l e t o c o m p e t e m o s t > s u c c e s s f u l l y a n d d e v e l o p s t r o n g s y n a p s > e s , w h i l e s y n a p s e s o f l o n g e r - l a t e n c y o r > l e s s - e f f e c t i v e i n p u t s a r e w e a k e n e d . > - A N D - > > 5 b . S o n g S , A b b o t t L F . N e u r o n . C o r t i c a l > d e v e l o p m e n t a n d r e m a p p i n g t h r o u g h s p i k > e t i m i n g - d e p e n d e n t p l a s t i c i t y . 2 0 0 1 O c t > 2 5 ; 3 2 ( 2 ) : 3 3 9 - 5 0 A b s t r a c t . L o n g - t e r m m o d > i f i c a t i o n o f s y n a p t i c e f f i c a c y c a n d e p e > n d o n t h e t i m i n g o f p r e - a n d p o s t s y n a p t > i c a c t i o n p o t e n t i a l s . I n m o d e l s t u d i e s , > s u c h s p i k e t i m i n g - d e p e n d e n t p l a s t i c i t y > ( S T D P ) i n t r o d u c e s t h e d e s i r a b l e f e a t u r > e s o f c o m p e t i t i o n a m o n g s y n a p s e s a n d r e > g u l a t i o n o f p o s t s y n a p t i c f i r i n g c h a r a c t > e r i s t i c s . S T D P s t r e n g t h e n s s y n a p s e s t h a > t r e c e i v e c o r r e l a t e d i n p u t , w h i c h c a n l > e a d t o t h e f o r m a t i o n o f s t i m u l u s - s e l e c t > i v e c o l u m n s a n d t h e d e v e l o p m e n t , r e f i n e > m e n t , a n d m a i n t e n a n c e o f s e l e c t i v i t y m a > p s i n n e t w o r k m o d e l s . T h e t e m p o r a l a s y m > m e t r y o f S T D P s u p p r e s s e s s t r o n g d e s t a b i > l i z i n g s e l f - e x c i t a t o r y l o o p s a n d a l l o w s > a g r o u p o f n e u r o n s t h a t b e c o m e s e l e c t i > v e e a r l y i n d e v e l o p m e n t t o d i r e c t o t h e r > n e u r o n s t o b e c o m e s i m i l a r l y s e l e c t i v e . > S T D P , a c t i n g a l o n e w i t h o u t f u r t h e r h y p o > t h e t i c a l g l o b a l c o n s t r a i n t s o r a d d i t i o n > a l f o r m s o f p l a s t i c i t y , c a n a l s o r e p r o d > u c e t h e r e m a p p i n g s e e n i n a d u l t c o r t e x > f o l l o w i n g a f f e r e n t l e s i o n s . T h e p a p e r s > a b o v e h a v e b e e n s e m i n a l i n i l l u s t r a t i n > g t h e i m p l i c a t i o n s f o r l e a r n i n g o f s p i k e > - t i m i n g - d e p e n d e n t s y n a p t i c p l a s t i c i t y > 6 . P o l s k y A , M e l B W , S c h i l l e r J . N a t N e > u r o s c i . 2 0 0 4 J u n ; 7 ( 6 ) : 6 2 1 - 7 . E p u b 2 0 0 4 > M a y 2 3 . C o m p u t a t i o n a l s u b u n i t s i n t h i n d > e n d r i t e s o f p y r a m i d a l c e l l s . A b s t r a c t . T > h e t h i n b a s a l a n d o b l i q u e d e n d r i t e s o f > c o r t i c a l p y r a m i d a l n e u r o n s r e c e i v e m o s > t o f t h e s y n a p t i c i n p u t s f r o m o t h e r c e l > l s , b u t t h e i r i n t e g r a t i v e p r o p e r t i e s r e > m a i n u n c e r t a i n . P r e v i o u s s t u d i e s h a v e m > o s t o f t e n r e p o r t e d g l o b a l l i n e a r o r s u b > l i n e a r s u m m a t i o n . A n a l t e r n a t i v e v i e w , > s u p p o r t e d b y b i o p h y s i c a l m o d e l i n g s t u d > i e s , h o l d s t h a t t h i n d e n d r i t e s p r o v i d e > a l a y e r o f i n d e p e n d e n t c o m p u t a t i o n a l ' s > u b u n i t s ' t h a t s i g m o i d a l l y m o d u l a t e t h e i > r i n p u t s p r i o r t o g l o b a l s u m m a t i o n . T o d > i s t i n g u i s h t h e s e p o s s i b i l i t i e s , w e c o m b > i n e d c o n f o c a l i m a g i n g a n d d u a l - s i t e f o c > a l s y n a p t i c s t i m u l a t i o n o f i d e n t i f i e d t > h i n d e n d r i t e s i n r a t n e o c o r t i c a l p y r a m i > d a l n e u r o n s . W e f o u n d t h a t n e a r b y i n p u t > s o n t h e s a m e b r a n c h s u m m e d s i g m o i d a l l y > , w h e r e a s w i d e l y s e p a r a t e d i n p u t s o r i n > p u t s t o d i f f e r e n t b r a n c h e s s u m m e d l i n e a > r l y . T h i s s t r o n g s p a t i a l c o m p a r t m e n t a l i > z a t i o n e f f e c t i s i n c o m p a t i b l e w i t h a g l > o b a l s u m m a t i o n r u l e a n d p r o v i d e s t h e f i > r s t e x p e r i m e n t a l s u p p o r t f o r a t w o - l a y e > r ' n e u r a l n e t w o r k ' m o d e l o f p y r a m i d a l n > e u r o n t h i n - b r a n c h i n t e g r a t i o n . O u r f i n d > i n g s c o u l d h a v e i m p o r t a n t i m p l i c a t i o n s > f o r t h e c o m p u t i n g a n d m e m o r y - r e l a t e d f > u n c t i o n s o f c o r t i c a l t i s s u e . T h i s p a p e r , > a s w e l l a s p r e v i o u s t h e o r e t i c a l w o r k , > s u g g e s t s t h a t d e n d r i t e s m i g h t e n a b l e s i > n g l e n e u r o n s t o b e h a v e a s f e e d f o r w a r d n > e u r a l n e t w o r k s . > 7 . M e d i n a J F , N o r e s W L , M a u k M D . N a t u r > e . 2 0 0 2 M a r 2 1 ; 4 1 6 ( 6 8 7 8 ) : 3 3 0 - 3 . I n h i b i t i > o n o f c l i m b i n g f i b r e s i s a s i g n a l f o r t > h e e x t i n c t i o n o f c o n d i t i o n e d e y e l i d r e s > p o n s e s . A b s t r a c t . A f u n d a m e n t a l t e n e t o f > c e r e b e l l a r l e a r n i n g t h e o r i e s a s s e r t s t > h a t c l i m b i n g f i b r e a f f e r e n t s f r o m t h e i > n f e r i o r o l i v e p r o v i d e a t e a c h i n g s i g n a l > t h a t p r o m o t e s t h e g r a d u a l a d a p t a t i o n o > f m o v e m e n t s . D a t a f r o m s e v e r a l f o r m s o f > m o t o r l e a r n i n g p r o v i d e s u p p o r t f o r t h i > s t e n e t . I n p a v l o v i a n e y e l i d c o n d i t i o n i > n g , f o r e x a m p l e , w h e r e a t o n e i s r e p e a t > e d l y p a i r e d w i t h a r e i n f o r c i n g u n c o n d i t > i o n e d s t i m u l u s l i k e p e r i o r b i t a l s t i m u l a > t i o n , t h e u n c o n d i t i o n e d s t i m u l u s p r o m o t > e s a c q u i s i t i o n o f c o n d i t i o n e d e y e l i d r e > s p o n s e s b y a c t i v a t i n g c l i m b i n g f i b r e s . > C l i m b i n g f i b r e a c t i v i t y e l i c i t e d b y a n > u n c o n d i t i o n e d s t i m u l u s i s i n h i b i t e d d u > r i n g t h e e x p r e s s i o n o f c o n d i t i o n e d r e s p o > n s e s - c o n s i s t e n t w i t h t h e i n h i b i t o r y p r o > j e c t i o n f r o m t h e c e r e b e l l u m t o i n f e r i o r > o l i v e . H e r e , w e s h o w t h a t i n h i b i t i o n o > f c l i m b i n g f i b r e s s e r v e s a s a t e a c h i n g > s i g n a l f o r e x t i n c t i o n , w h e r e l e a r n i n g n > o t t o r e s p o n d i s s i g n a l l e d b y p r e s e n t i n > g a t o n e w i t h o u t t h e u n c o n d i t i o n e d s t i m > u l u s . W e u s e d r e v e r s i b l e i n f u s i o n o f s y > n a p t i c r e c e p t o r a n t a g o n i s t s t o s h o w t h a > t b l o c k i n g i n h i b i t o r y i n p u t t o t h e c l i m > b i n g f i b r e s p r e v e n t s e x t i n c t i o n o f t h e > c o n d i t i o n e d r e s p o n s e , w h e r e a s b l o c k i n g > e x c i t a t o r y i n p u t i n d u c e s e x t i n c t i o n . T > h e s e r e s u l t s , c o m b i n e d w i t h a n a l y s i s o f > c l i m b i n g f i b r e a c t i v i t y i n a c o m p u t e r > s i m u l a t i o n o f t h e c e r e b e l l a r - o l i v a r y s y > s t e m , s u g g e s t t h a t t r a n s i e n t i n h i b i t i o n > o f c l i m b i n g f i b r e s b e l o w t h e i r b a c k g r o > u n d l e v e l i s t h e s i g n a l t h a t d r i v e s e x t > i n c t i o n . T h i s i s o n e o f s e v e r a l c o m p u t a t > i o n a l s t u d i e s b y M a u k a n d c o l l a b o r a t o r s > t h a t a r e e n h a n c i n g o u r k n o w l e d g e o f c e > r e b e l l a r p r o c e s s i n g ( a l s o s e e s i m i l a r p > a p e r s b y R a y m o n d & L i s b e r g e r a p p l i e d t o > t h e V O R ) . > 8 . O l s h a u s e n B A , F i e l d D J . N a t u r e . E m > e r g e n c e o f s i m p l e - c e l l r e c e p t i v e f i e l d > p r o p e r t i e s b y l e a r n i n g a s p a r s e c o d e f o > r n a t u r a l i m a g e s . 1 9 9 6 J u n 1 3 ; 3 8 1 ( 6 5 8 3 ) > : 6 0 7 - 9 A b s t r a c t . T h e r e c e p t i v e f i e l d s o f > s i m p l e c e l l s i n m a m m a l i a n p r i m a r y v i s u a > l c o r t e x c a n b e c h a r a c t e r i z e d a s b e i n g > s p a t i a l l y l o c a l i z e d , o r i e n t e d a n d b a n d > p a s s ( s e l e c t i v e t o s t r u c t u r e a t d i f f e r e > n t s p a t i a l s c a l e s ) , c o m p a r a b l e t o t h e b > a s i s f u n c t i o n s o f w a v e l e t t r a n s f o r m s . O > n e a p p r o a c h t o u n d e r s t a n d i n g s u c h r e s p o n > s e p r o p e r t i e s o f v i s u a l n e u r o n s h a s b e e > n t o c o n s i d e r t h e i r r e l a t i o n s h i p t o t h e > s t a t i s t i c a l s t r u c t u r e o f n a t u r a l i m a g e > s i n t e r m s o f e f f i c i e n t c o d i n g . A l o n g t > h e s e l i n e s , a n u m b e r o f s t u d i e s h a v e a t > t e m p t e d t o t r a i n u n s u p e r v i s e d l e a r n i n g > a l g o r i t h m s o n n a t u r a l i m a g e s i n t h e h o > p e o f d e v e l o p i n g r e c e p t i v e f i e l d s w i t h > s i m i l a r p r o p e r t i e s , b u t n o n e h a s s u c c e e > d e d i n p r o d u c i n g a f u l l s e t t h a t s p a n s > t h e i m a g e s p a c e a n d c o n t a i n s a l l t h r e e > o f t h e a b o v e p r o p e r t i e s . H e r e w e i n v e s t > i g a t e t h e p r o p o s a l t h a t a c o d i n g s t r a t e > g y t h a t m a x i m i z e s s p a r s e n e s s i s s u f f i c i > e n t t o a c c o u n t f o r t h e s e p r o p e r t i e s . W e > s h o w t h a t a l e a r n i n g a l g o r i t h m t h a t a t > t e m p t s t o f i n d s p a r s e l i n e a r c o d e s f o r > n a t u r a l s c e n e s w i l l d e v e l o p a c o m p l e t e > f a m i l y o f l o c a l i z e d , o r i e n t e d , b a n d p a s > s r e c e p t i v e f i e l d s , s i m i l a r t o t h o s e f o > u n d i n t h e p r i m a r y v i s u a l c o r t e x . T h e r > e s u l t i n g s p a r s e i m a g e c o d e p r o v i d e s a m > o r e e f f i c i e n t r e p r e s e n t a t i o n f o r l a t e r > s t a g e s o f p r o c e s s i n g b e c a u s e i t p o s s e s s > e s a h i g h e r d e g r e e o f s t a t i s t i c a l i n d e p > e n d e n c e a m o n g i t s o u t p u t s . T h i s n o w c l a s > s i c s t u d y s u g g e s t s h o w t h e s t a t i s t i c a l > s t r u c t u r e o f n a t u r a l i m a g e s m a y d e t e r m i > n e t h e r e s p o n s e p r o p e r t i e s o f V 1 c e l l s , > a n d s e t t h e s t a g e f o r m a n y l a t e r s t u d i > e s d i s c u s s i n g t h e c o n c e p t o f s p a r s e c o > d i n g o f i m a g e s . > 9 . T a y l o r D M , T i l l e r y S I , S c h w a r t z A B . > D i r e c t c o r t i c a l c o n t r o l o f 3 D n e u r o p r o > s t h e t i c d e v i c e s . S c i e n c e . 2 0 0 2 J u n 7 ; 2 9 > 6 ( 5 5 7 4 ) : 1 8 2 9 - 3 2 . A b s t r a c t . T h r e e - d i m e n s > i o n a l ( 3 D ) m o v e m e n t o f n e u r o p r o s t h e t i c > d e v i c e s c a n b e c o n t r o l l e d b y t h e a c t i v i > t y o f c o r t i c a l n e u r o n s w h e n a p p r o p r i a t e > a l g o r i t h m s a r e u s e d t o d e c o d e i n t e n d e d > m o v e m e n t i n r e a l t i m e . P r e v i o u s s t u d i e > s a s s u m e d t h a t n e u r o n s m a i n t a i n f i x e d t > u n i n g p r o p e r t i e s , a n d t h e s t u d i e s u s e d > s u b j e c t s w h o w e r e u n a w a r e o f t h e m o v e m > e n t s p r e d i c t e d b y t h e i r r e c o r d e d u n i t s . > I n t h i s s t u d y , s u b j e c t s h a d r e a l - t i m e > v i s u a l f e e d b a c k o f t h e i r b r a i n - c o n t r o l l > e d t r a j e c t o r i e s . C e l l t u n i n g p r o p e r t i e s > c h a n g e d w h e n u s e d f o r b r a i n - c o n t r o l l e d > m o v e m e n t s . B y u s i n g c o n t r o l a l g o r i t h m s > t h a t t r a c k t h e s e c h a n g e s , s u b j e c t s m a d > e l o n g s e q u e n c e s o f 3 D m o v e m e n t s u s i n g > f a r f e w e r c o r t i c a l u n i t s t h a n e x p e c t e d . > D a i l y p r a c t i c e i m p r o v e d m o v e m e n t a c c u r > a c y a n d t h e d i r e c t i o n a l t u n i n g o f t h e s e > u n i t s . T h i s r e p r e s e n t s s o m e o f t h e s e m i > n a l w o r k d e c o d i n g c o r t i c a l a c t i v i t y t o > c o n t r o l n e u r a l p r o s t h e t i c s . > 1 0 . V a n V r e e s w i j k C , A b b o t t L F , E r m e n t > r o u t G B . W h e n i n h i b i t i o n n o t e x c i t a t i o n > s y n c h r o n i z e s n e u r a l f i r i n g . J C o m p u t N > e u r o s c i . 1 9 9 4 D e c ; 1 ( 4 ) : 3 1 3 - 2 1 . A b s t r a c t . > E x c i t a t o r y a n d i n h i b i t o r y s y n a p t i c c o u > p l i n g c a n h a v e c o u n t e r - i n t u i t i v e e f f e c t > s o n t h e s y n c h r o n i z a t i o n o f n e u r o n a l f i > r i n g . W h i l e i t m i g h t a p p e a r t h a t e x c i t a > t o r y c o u p l i n g w o u l d l e a d t o s y n c h r o n i z a > t i o n , w e s h o w t h a t f r e q u e n t l y i n h i b i t i o > n r a t h e r t h a n e x c i t a t i o n s y n c h r o n i z e s f > i r i n g . W e s t u d y t w o i d e n t i c a l n e u r o n s d > e s c r i b e d b y i n t e g r a t e - a n d - f i r e m o d e l s , > g e n e r a l p h a s e - c o u p l e d m o d e l s o r t h e H o d > g k i n - H u x l e y m o d e l w i t h m u t u a l , n o n - i n s t > a n t a n e o u s e x c i t a t o r y o r i n h i b i t o r y s y n a > p s e s b e t w e e n t h e m . W e f i n d t h a t i f t h e > r i s e t i m e o f t h e s y n a p s e i s l o n g e r t h a n > t h e d u r a t i o n o f a n a c t i o n p o t e n t i a l , i > n h i b i t i o n n o t e x c i t a t i o n l e a d s t o s y n c h > r o n i z e d f i r i n g . > I f I w e r e t o u p d a t e , I t h i n k I w o u l d a > d d p a p e r s f r o m : > 1 ) N e u r o e c o n o m i c s & R e i n f o r c e m e n t L e a r > n i n g - - i n a d d i t i o n t o t h e s e m i n a l w o r k > b y D a y a n & S c h u l t z ( a l r e a d y i n a t t a c h e d > ) , p e r h a p s L o e w e n s t e i n / S e u n g p a p e r o n m > a t c h i n g b e h a v i o r a s a g e n e r i c c o n s e q u e n > c e o f c o r r e l a t i o n a l l e a r n i n g r u l e s . > 2 ) B a y e s i a n n e t w o r k s - - m a y b e M a , B e c k > , L a t h a m , P o u g e t o r o t h e r s o n i d e a t h a t > t h e b r a i n m a y e n c o d e & c o m p u t e w i t h p r > o b a b i l i t i e s > - - END OF FILE > -- > -- > Dr Jim Stone, > Psychology Department, Sheffield University, Sheffield, S10 2TP, UK. > Tel: 0114 2226522. http://jim-stone.staff.shef.ac.uk/ > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080718/e9ef2602/attachment-0001.html From bower at uthscsa.edu Tue Jul 22 12:46:25 2008 From: bower at uthscsa.edu (jim bower) Date: Tue Jul 22 13:11:21 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <002701c8ebdd$eb4b9cc0$81c9f889@braun03> References: <20080718111439.4E18A8FEE3C@neuroinf.org><48808F72.80501@cmu.edu> <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry><002701c8ebdd$eb4b9cc0$81c9f889@braun03> Message-ID: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> One other general point about oscillations. Years ago a "neural network" engineer from MIT gave a talk at the Snowbird meeting, I think the first - in his talk he that, connected at random only 1 percent of networks didn't oscillate intrinsically, and he proposed to find those networks as they were clearly the only ones that were useful. Interesting idea but dead wrong. Everything in biology oscillates, in fact everything in the natural world does. Engineers fear oscillations because they don't know how to control them. The nervous system uses them to its own purposes. In fact, my guess is that this is one of the sources of its efficiency. Last point with respect to your car, the quality of the engineer must be based on the performance of what it has built. So last last question, does anyone know something whose perfomance is more extraordianary then the brain of a fly? Or a slug? I don't think so Jim Sent via BlackBerry by AT&T -----Original Message----- From: "Hans A. Braun" Date: Tue, 22 Jul 2008 11:32:54 To: ; Nathan Urban; ; Subject: Re: [Comp-neuro] Review announcement Hi Jim, nice to hear from you with an interesting question. Here is a question back: Who says that the biological coding scheme is optimised in a way as engineers would do? I have been educated as an engineer. Thereby, I specifically have learnt how handle, if it cannot be avoided, such detrimental system properties like noise, nonlinearities and time delays because these can lead to unpredictable system behavior, including undesired oscillation and chaos ? what regularly can be seen in all kind of biological systems. If something similar would happen in a car or an airplane, the responsible engineer, deservedly, would immediately be fired. Could it be that the engineer in evolution has used a principally different strategy? What was/is his/her goal? Who knows or who is interested to find the answer? - or a more appropriate question ;-) ? Coming back to the original "noise" question: During all the years as experimental physiologist I have got hundreds of hours recordings of impulse sequences from different neurons ? and all look more or less noisy - whatever it means. Best wishes Hans Braun PS: if you are interested, here are two references to our work (an actual and an earlier paper): Finke C, Vollmer J, Postnova S, Braun HA (2008) Propagation effects of current and conductance noise in a model neuron with subthreshold oscillations. Mathematical Biosciences doi:10.1016/j.mbs.2008.03.007 Braun HA, Wissing H, Sch?fer K, Hirsch MC (1994). Oscillation and noise determine signal transduction in shark multimodal sensory cells. Nature 367: 270-273. The first one is a mathematical/computational approach which has very recently been published, so far only as online version. The second reference is to a much earlier experimental paper which demonstrates how the evolutionary engineer might have used oscillations and noise to achieve a particular sensitivity. This strategy, for whatever reasons, was only realized for sensory encoding in some evolutionary very old animals like sharks. Dr. Hans A. Braun, Institute of Physiology, Deutschhausstr. 2, D-35037 Marburg, Germany. Tel: +49 (0)6421-286 23 05, FAX: +49 (0)6421-286 6967, E-mail: braun@staff.uni-marburg.de URL: http://www.uni-marburg.de/physiology/braun and http://www.clabs.de see also: http://www.BioSim-Network.net ----- Original Message ----- From: "jim bower" To: "Nathan Urban" ; ; Sent: Friday, July 18, 2008 3:56 PM Subject: Re: [Comp-neuro] Review announcement > Haven't done this in a long time. But who says neurons are noisy? > > From the point of view of information theory, why isn't the apperance of noise expected in a highly optimized coding scheme? And why isn't synchrony to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > > Just curious. > > Jim bower > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Nathan Urban > > Date: Fri, 18 Jul 2008 08:41:22 > To: > Subject: [Comp-neuro] Review announcement > > > Review announcement > > This review describes a constructive role for noise in synchronizing > populations of neurons and should be of interest to computaional > neurosciuentists. > > > Trends Neurosci. 2008 Jul 4. [Epub ahead of print] > Reliability, synchrony and noise. > Ermentrout GB, Gal?n RF, Urban NN. > > The brain is noisy. Neurons receive tens of thousands of highly > fluctuating inputs and generate spike trains that appear highly > irregular. Much of this activity is spontaneous - uncoupled to overt > stimuli or motor outputs - leading to questions about the functional > impact of this noise. Although noise is most often thought of as > disrupting patterned activity and interfering with the encoding of > stimuli, recent theoretical and experimental work has shown that noise > can play a constructive role - leading to increased reliability or > regularity of neuronal firing in single neurons and across populations. > These results raise fundamental questions about how noise can influence > neural function and computation. > > PMID: 18603311 [PubMed - as supplied by publisher] > > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_us er=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_versio n=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > ---------------------------------------------------------------------------- ---- > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > From aaf23 at cam.ac.uk Tue Jul 22 13:13:26 2008 From: aaf23 at cam.ac.uk (Aldo Faisal) Date: Tue Jul 22 15:10:10 2008 Subject: [Comp-neuro] Re: Review announcement Message-ID: > Haven't done this in a long time. But who says neurons are noisy? I am much enjoying this discussion! Whether the observed neuronal variability is due to true randomness (i.e. noise) or something that just looks like noise is indeed not knew. However, what has changed is that we can now - from bottom-up - understand how randomness from molecular processes (I assume here diffusion, chemical reactions, fluctuations in signalling proteins confirmation are stochastic) influences electrical activity of whole cell behaviour and how this can produce variability in (cellular) behaviour accounting for much of the overall variability [1]. While in a large neuron these microscopic noise sources are negligible, their effects becomes quite significant in the miniaturized (and thus energy efficient) circuits of our cortex. In fact noise sets a universal lower limit to the size of function neurons that is matched by anatomical data across species [2] and will produce steadily increasing variability in propagating action potentials [3]. Thus, noise from molecular sources can have an impact in the highly energy efficient and thus compact circuits of our brains, setting lower limits to how precise we can encode information. How to design (like an engineer) a brain's circuits is thus a multi-factor trade-off problem...and to me quite fascinating. [1] Faisal, A.A., Selen, L.P. & Wolpert, D.M. (2008) "Noise in the nervous system". Nat. Rev. Neurosci. 9, 292?303 [2] Faisal,A.A., White,J.A. & Laughlin, S.B. (2005) "Ion channel noise places limits to the miniaturization of the brain's wiring", Curr Biol, Vol. 15(12), pp. 1143-1149 [3] Faisal, A.A. and Laughlin, S.B. (2007) "Stochastic simulations on the reliability of action potential propagation in thin axons", PLOS Comp. Biol. 3(5):e79 From minai_ali at yahoo.com Tue Jul 22 14:57:21 2008 From: minai_ali at yahoo.com (Ali Minai) Date: Tue Jul 22 15:10:14 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> Message-ID: <318893.22356.qm@web65511.mail.ac4.yahoo.com> Nature clearly uses a different "engineering" strategy than humans. Classical human engineering is based on control (controlling dynamics, controlling noise, controlling quality, etc.) and is primarily goal-oriented. Thus, its capabilities are limited by the imaginations of those setting the goals. Things like noise, oscillation and combinatorial richness interfere with this imagination (such as it is) and are thus seen as hazards. Nature's engineering, in contrast, is based on exploitation (exploiting oscillations, exploiting noise, exploiting variation, exploiting chance combinations, etc.), and its capabilities are limited only by the possibilities offered by the phenomena at hand. As this natural engineering configures more complex phenomena, the space of possibilities *expands*, thus making even more complex phenomena possible. Thus, Nature's engineering is open-ended, and things such as noise, variation and combinatorial richness are seen as enablers rather than problems. To be fair, human engineering has very different purposes than Nature, and its approach is well-suited to those purposes, but I think we're starting to build things where Nature's way might be the only option. The accompanying loss of control is, of course, inevitable. Ali --------------------------------------------------------------------- Ali A. Minai Associate Professor Associate Head for Electrical Engineering Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: aminai@ececs.uc.edu ????????? minai_ali@yahoo.com WWW: http://www.ececs.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Tue, 7/22/08, jim bower wrote: From: jim bower Subject: Re: [Comp-neuro] Review announcement To: "Hans A. Braun" , "Nathan Urban" , comp-neuro-bounces@neuroinf.org, comp-neuro@neuroinf.org Date: Tuesday, July 22, 2008, 6:46 AM One other general point about oscillations. Years ago a "neural network" engineer from MIT gave a talk at the Snowbird meeting, I think the first - in his talk he that, connected at random only 1 percent of networks didn't oscillate intrinsically, and he proposed to find those networks as they were clearly the only ones that were useful. Interesting idea but dead wrong. Everything in biology oscillates, in fact everything in the natural world does. Engineers fear oscillations because they don't know how to control them. The nervous system uses them to its own purposes. In fact, my guess is that this is one of the sources of its efficiency. Last point with respect to your car, the quality of the engineer must be based on the performance of what it has built. So last last question, does anyone know something whose perfomance is more extraordianary then the brain of a fly? Or a slug? I don't think so Jim Sent via BlackBerry by AT&T -----Original Message----- From: "Hans A. Braun" Date: Tue, 22 Jul 2008 11:32:54 To: ; Nathan Urban; ; Subject: Re: [Comp-neuro] Review announcement Hi Jim, nice to hear from you with an interesting question. Here is a question back: Who says that the biological coding scheme is optimised in a way as engineers would do? I have been educated as an engineer. Thereby, I specifically have learnt how handle, if it cannot be avoided, such detrimental system properties like noise, nonlinearities and time delays because these can lead to unpredictable system behavior, including undesired oscillation and chaos ? what regularly can be seen in all kind of biological systems. If something similar would happen in a car or an airplane, the responsible engineer, deservedly, would immediately be fired. Could it be that the engineer in evolution has used a principally different strategy? What was/is his/her goal? Who knows or who is interested to find the answer? - or a more appropriate question ;-) ? Coming back to the original "noise" question: During all the years as experimental physiologist I have got hundreds of hours recordings of impulse sequences from different neurons ? and all look more or less noisy - whatever it means. Best wishes Hans Braun PS: if you are interested, here are two references to our work (an actual and an earlier paper): Finke C, Vollmer J, Postnova S, Braun HA (2008) Propagation effects of current and conductance noise in a model neuron with subthreshold oscillations. Mathematical Biosciences doi:10.1016/j.mbs.2008.03.007 Braun HA, Wissing H, Sch?fer K, Hirsch MC (1994). Oscillation and noise determine signal transduction in shark multimodal sensory cells. Nature 367: 270-273. The first one is a mathematical/computational approach which has very recently been published, so far only as online version. The second reference is to a much earlier experimental paper which demonstrates how the evolutionary engineer might have used oscillations and noise to achieve a particular sensitivity. This strategy, for whatever reasons, was only realized for sensory encoding in some evolutionary very old animals like sharks. Dr. Hans A. Braun, Institute of Physiology, Deutschhausstr. 2, D-35037 Marburg, Germany. Tel: +49 (0)6421-286 23 05, FAX: +49 (0)6421-286 6967, E-mail: braun@staff.uni-marburg.de URL: http://www.uni-marburg.de/physiology/braun and http://www.clabs.de see also: http://www.BioSim-Network.net ----- Original Message ----- From: "jim bower" To: "Nathan Urban" ; ; Sent: Friday, July 18, 2008 3:56 PM Subject: Re: [Comp-neuro] Review announcement > Haven't done this in a long time. But who says neurons are noisy? > > From the point of view of information theory, why isn't the apperance of noise expected in a highly optimized coding scheme? And why isn't synchrony to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > > Just curious. > > Jim bower > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Nathan Urban > > Date: Fri, 18 Jul 2008 08:41:22 > To: > Subject: [Comp-neuro] Review announcement > > > Review announcement > > This review describes a constructive role for noise in synchronizing > populations of neurons and should be of interest to computaional > neurosciuentists. > > > Trends Neurosci. 2008 Jul 4. [Epub ahead of print] > Reliability, synchrony and noise. > Ermentrout GB, Gal?n RF, Urban NN. > > The brain is noisy. Neurons receive tens of thousands of highly > fluctuating inputs and generate spike trains that appear highly > irregular. Much of this activity is spontaneous - uncoupled to overt > stimuli or motor outputs - leading to questions about the functional > impact of this noise. Although noise is most often thought of as > disrupting patterned activity and interfering with the encoding of > stimuli, recent theoretical and experimental work has shown that noise > can play a constructive role - leading to increased reliability or > regularity of neuronal firing in single neurons and across populations. > These results raise fundamental questions about how noise can influence > neural function and computation. > > PMID: 18603311 [PubMed - as supplied by publisher] > > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_us er=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_versio n=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > ---------------------------------------------------------------------------- ---- > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro >_______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/06f363e1/attachment-0001.html From nurban at cmu.edu Tue Jul 22 14:55:39 2008 From: nurban at cmu.edu (Nathan Urban) Date: Tue Jul 22 15:10:54 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> References: <20080718111439.4E18A8FEE3C@neuroinf.org><48808F72.80501@cmu.edu> <931100367-1216389428-cardhu_decombobulator_blackberry.rim.net-1792876130-@bxe111.bisx.prod.on.blackberry><002701c8ebdd$eb4b9cc0$81c9f889@braun03> <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> Message-ID: <4885D8CB.6040603@cmu.edu> All, I am glad for all the interesting comments. A few points in response. 1) While I am skeptical of the claim that only 1% of networks don't oscillate, even if we take it as true, one of the key situations that we are discussing in this review is the case of unconnected networks that receive common aperiodic input. My own intuitions were strongly against the idea that synchronous oscillations should arise in these scenarios. However, we showed experimentally (Galan et al 2006) that this occurs in neurons under certain conditions and then have explored this in simulations and in theory. Perhaps not surprisingly similar phenomena were already known in theoretical physics from the work of Pikovsky and Rosenblum and others in the 1980s on the synchronization of chaotic systems by common inputs. Such noise-induced synchrony is facilitated when neurons are good oscillators (like Hodgkin and Huxley's squid axons) but still works when neurons are not very robust oscillators - like leakyintegrate and fire neurons. This general phenomenon of correlated noise-induced oscillatory synchrony is quite robust and has been observed or proposed in some other biological systems. 2) Is the brain "noisy" and is this surprising? Of course this depends on what you define as "noise". For the review we defined "noise" as behaving as if drawn randomly from a distribution. Thus, the fact that membrane current measured in vivo is gaussian distributed and the fact that spike trains are often well approximated by a poisson process is what we mean by "noisy" and if the distribution from which the signal is drawn has a large variance, then this is more noisy. This does not mean that such signals are unreliable. In fact, experiments like those of Bryant and Segundo and Mainen and Sejnowski (and our own Galan et al 2008) point out that more noise induces more reliability, either across trials or across cells in the same trial. 3) I think that an interesting question that arises from this discussion is how to code efficiently with strongly oscillating neurons. Very efficient coding will indeed, as Jim suggests, often look like noise. However, various networks in the brain oscillate, and these oscillations often occur when we think that part of the brain is active and functioning (hippocampal theta oscillations, olfactory bulb gamma oscillations). Given that the regularity of oscillators reduces the capacity to represent information, what is the best that we can do in representing a lot of information when our favorite brain area is functioning in a way that is arguably sub-optimal (by oscillating). 4) People certainly have highlighted the idea that noise can be good for at least 30 years, and in fact probably a lot longer than this. The original work on stochastic resonance was in teh late 1970's, I believe. However, I think that biologists and even physiologists sometimes fail to appreciate the constructive effects of noise. (E.g. see Dan Wolpert's recent review on Noise ion teh Nervous System in Nature Reviews Neuroscience.) Also, there are new examples that are being added to the repertoire of such constructive effects noise,a dn tehse phenomena are being more clearly understood. In the review we focus on this particular example of noise as facilitating synchronization of oscillations, drawing form our own experimental work and that of others, and also on making the parallel between synchrony and reliability. Well, enough for now. I hope that everyone reads and enjoys the review Nathan jim bower wrote: > One other general point about oscillations. Years ago a "neural network" engineer from MIT gave a talk at the Snowbird meeting, I think the first - in his talk he that, connected at random only 1 percent of networks didn't oscillate intrinsically, and he proposed to find those networks as they were clearly the only ones that were useful. > > Interesting idea but dead wrong. Everything in biology oscillates, in fact everything in the natural world does. Engineers fear oscillations because they don't know how to control them. The nervous system uses them to its own purposes. In fact, my guess is that this is one of the sources of its efficiency. > > Last point with respect to your car, the quality of the engineer must be based on the performance of what it has built. So last last question, does anyone know something whose perfomance is more extraordianary then the brain of a fly? Or a slug? > > I don't think so > > Jim > > > Sent via BlackBerry by AT&T > > -----Original Message----- > From: "Hans A. Braun" > > Date: Tue, 22 Jul 2008 11:32:54 > To: ; Nathan Urban; ; > Subject: Re: [Comp-neuro] Review announcement > > > Hi Jim, > > > > nice to hear from you with an interesting question. Here is a question back: > > Who says that the biological coding scheme is optimised in a way as > engineers would do? > > > > I have been educated as an engineer. Thereby, I specifically have learnt how > handle, if it cannot be avoided, such detrimental system properties like > noise, nonlinearities and time delays because these can lead to > unpredictable system behavior, including undesired oscillation and chaos ? > what regularly can be seen in all kind of biological systems. If something > similar would happen in a car or an airplane, the responsible engineer, > deservedly, would immediately be fired. > > > > Could it be that the engineer in evolution has used a principally different > strategy? > > What was/is his/her goal? > > Who knows or who is interested to find the answer? > > - or a more appropriate question ;-) ? > > > > Coming back to the original "noise" question: During all the years as > experimental physiologist I have got hundreds of hours recordings of impulse > sequences from different neurons ? and all look more or less noisy - > whatever it means. > > > > Best wishes > > Hans Braun > > > > PS: if you are interested, here are two references to our work (an actual > and an earlier paper): > > > > Finke C, Vollmer J, Postnova S, Braun HA (2008) Propagation effects of > current and conductance noise in a model neuron with subthreshold > oscillations. Mathematical Biosciences doi:10.1016/j.mbs.2008.03.007 > > Braun HA, Wissing H, Sch?fer K, Hirsch MC (1994). Oscillation and noise > determine signal transduction in shark multimodal sensory cells. Nature 367: > 270-273. > > > > The first one is a mathematical/computational approach which has very > recently been published, so far only as online version. > > The second reference is to a much earlier experimental paper which > demonstrates how the evolutionary engineer might have used oscillations and > noise to achieve a particular sensitivity. This strategy, for whatever > reasons, was only realized for sensory encoding in some evolutionary very > old animals like sharks. > > Dr. Hans A. Braun, Institute of Physiology, Deutschhausstr. 2, D-35037 > Marburg, Germany. > Tel: +49 (0)6421-286 23 05, FAX: +49 (0)6421-286 6967, E-mail: > braun@staff.uni-marburg.de > URL: http://www.uni-marburg.de/physiology/braun and http://www.clabs.de > see also: http://www.BioSim-Network.net > > ----- Original Message ----- > From: "jim bower" > To: "Nathan Urban" ; ; > > Sent: Friday, July 18, 2008 3:56 PM > Subject: Re: [Comp-neuro] Review announcement > > > >> Haven't done this in a long time. But who says neurons are noisy? >> >> From the point of view of information theory, why isn't the apperance of >> > noise expected in a highly optimized coding scheme? And why isn't synchrony > to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > >> Just curious. >> >> Jim bower >> Sent via BlackBerry by AT&T >> >> -----Original Message----- >> From: Nathan Urban >> >> Date: Fri, 18 Jul 2008 08:41:22 >> To: >> Subject: [Comp-neuro] Review announcement >> >> >> Review announcement >> >> This review describes a constructive role for noise in synchronizing >> populations of neurons and should be of interest to computaional >> neurosciuentists. >> >> >> Trends Neurosci. 2008 Jul 4. [Epub ahead of print] >> Reliability, synchrony and noise. >> Ermentrout GB, Gal?n RF, Urban NN. >> >> The brain is noisy. Neurons receive tens of thousands of highly >> fluctuating inputs and generate spike trains that appear highly >> irregular. Much of this activity is spontaneous - uncoupled to overt >> stimuli or motor outputs - leading to questions about the functional >> impact of this noise. Although noise is most often thought of as >> disrupting patterned activity and interfering with the encoding of >> stimuli, recent theoretical and experimental work has shown that noise >> can play a constructive role - leading to increased reliability or >> regularity of neuronal firing in single neurons and across populations. >> These results raise fundamental questions about how noise can influence >> neural function and computation. >> >> PMID: 18603311 [PubMed - as supplied by publisher] >> >> >> > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_us > er=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_versio > n=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > >> _______________________________________________ >> Comp-neuro mailing list >> Comp-neuro@neuroinf.org >> http://www.neuroinf.org/mailman/listinfo/comp-neuro >> >> > > > ---------------------------------------------------------------------------- > ---- > > > >> _______________________________________________ >> Comp-neuro mailing list >> Comp-neuro@neuroinf.org >> http://www.neuroinf.org/mailman/listinfo/comp-neuro >> >> > > -- ____________________________________________ Nathan Urban, Ph.D. Associate Professor Department of Biological Sciences and Center for the Neural Basis of Cognition Carnegie Mellon University 4400 Fifth Ave Pittsburgh PA 15213 ph. 412-268-5122 fax 412-268-8423 http://www.andrew.cmu.edu/user/nurban/Lab_pages/ From sabrahamse at gmail.com Tue Jul 22 15:49:56 2008 From: sabrahamse at gmail.com (Sven Abrahamse) Date: Tue Jul 22 15:55:19 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <318893.22356.qm@web65511.mail.ac4.yahoo.com> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> <318893.22356.qm@web65511.mail.ac4.yahoo.com> Message-ID: <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> what an amazing discussion this has turned out to be. Way to go guys. I am sure that everybody simply enjoys this format of things :-) On Tue, Jul 22, 2008 at 2:57 PM, Ali Minai wrote: > Nature clearly uses a different "engineering" strategy than humans. > Classical human engineering is based on control (controlling dynamics, > controlling noise, controlling quality, etc.) and is primarily > goal-oriented. Thus, its capabilities are limited by the imaginations of > those setting the goals. Things like noise, oscillation and combinatorial > richness interfere with this imagination (such as it is) and are thus seen > as hazards. Nature's engineering, in contrast, is based on exploitation > (exploiting oscillations, exploiting noise, exploiting variation, exploiting > chance combinations, etc.), and its capabilities are limited only by the > possibilities offered by the phenomena at hand. As this natural engineering > configures more complex phenomena, the space of possibilities *expands*, > thus making even more complex phenomena possible. Thus, Nature's engineering > is open-ended, and things such as noise, variation and combinatorial > richness are seen as enablers rather than problems. To be fair, human > engineering has very different purposes than Nature, and its approach is > well-suited to those purposes, but I think we're starting to build things > where Nature's way might be the only option. The accompanying loss of > control is, of course, inevitable. > > Ali > > > --------------------------------------------------------------------- > Ali A. Minai > Associate Professor > Associate Head for Electrical Engineering > Department of Electrical & Computer Engineering > University of Cincinnati > Cincinnati, OH 45221-0030 > > Phone: (513) 556-4783 > Fax: (513) 556-7326 > Email: aminai@ececs.uc.edu > minai_ali@yahoo.com > > WWW: http://www.ececs.uc.edu/~aminai/ > > ---------------------------------------------------------------------- > > --- On *Tue, 7/22/08, jim bower * wrote: > > From: jim bower > Subject: Re: [Comp-neuro] Review announcement > To: "Hans A. Braun" , "Nathan Urban" < > nurban@cmu.edu>, comp-neuro-bounces@neuroinf.org, comp-neuro@neuroinf.org > Date: Tuesday, July 22, 2008, 6:46 AM > > > One other general point about oscillations. Years ago a "neural > network" engineer from MIT gave a talk at the Snowbird meeting, I think > the first - in his talk he that, connected at random only 1 percent of networks > didn't oscillate intrinsically, and he proposed to find those networks as > they were clearly the only ones that were useful. > > Interesting > idea but dead wrong. Everything in biology oscillates, in fact > everything in the natural world does. Engineers fear oscillations because they > don't know how to control them. The nervous system uses them to its own > purposes. In fact, my guess is that this is one of the sources of its > efficiency. > > Last point with respect to your car, the quality of the engineer must be based > on the performance of what it has built. So last last question, does anyone > know something whose perfomance is more extraordianary then the brain of a fly? > Or a slug? > > I don't think so > > Jim > > > Sent via BlackBerry by AT&T > > -----Original Message----- > From: "Hans A. Braun" > > Date: Tue, 22 Jul 2008 11:32:54 > To: ; Nathan Urban; > ; > Subject: Re: [Comp-neuro] Review > announcement > > > Hi Jim, > > > > nice to hear from you with an interesting question. Here is a question back: > > Who says that the biological coding scheme is optimised in a way as > engineers would do? > > > > I have been educated as an engineer. Thereby, I specifically have learnt how > handle, if it cannot be avoided, such detrimental system properties like > noise, nonlinearities and time delays because these can lead to > unpredictable system behavior, including undesired oscillation and chaos ? > what regularly can be seen in all kind of biological systems. If something > similar would happen in a car or an airplane, the responsible engineer, > deservedly, would immediately be fired. > > > > Could it be that the engineer in evolution has used a principally different > strategy? > > What was/is his/her goal? > > Who knows or who is interested to find the answer? > > - or a more appropriate question > ;-) ? > > > > Coming back to the original "noise" question: During all the years > as > experimental physiologist I have got hundreds of hours recordings of impulse > sequences from different neurons ? and all look more or less noisy - > whatever it means. > > > > Best wishes > > Hans Braun > > > > PS: if you are interested, here are two references to our work (an actual > and an earlier paper): > > > > Finke C, Vollmer J, Postnova S, Braun HA (2008) Propagation effects of > current and conductance noise in a model neuron with subthreshold > oscillations. Mathematical Biosciences doi:10.1016/j.mbs.2008.03.007 > > Braun HA, Wissing H, Sch?fer K, Hirsch MC (1994). Oscillation and noise > determine signal transduction in shark multimodal sensory cells. Nature 367: > 270-273. > > > > The first one is a mathematical/computational approach which has very > recently been published, so far only as online > version. > > The second reference is to a much earlier experimental paper which > demonstrates how the evolutionary engineer might have used oscillations and > noise to achieve a particular sensitivity. This strategy, for whatever > reasons, was only realized for sensory encoding in some evolutionary very > old animals like sharks. > > Dr. Hans A. Braun, Institute of Physiology, Deutschhausstr. 2, D-35037 > Marburg, Germany. > Tel: +49 (0)6421-286 23 05, FAX: +49 (0)6421-286 6967, E-mail: > braun@staff.uni-marburg.de > URL: http://www.uni-marburg.de/physiology/braun and http://www.clabs.de > see also: http://www.BioSim-Network.net > > ----- Original Message ----- > From: "jim bower" > To: "Nathan Urban" ; > ; > > Sent: Friday, July 18, 2008 3:56 PM > Subject: Re: [Comp-neuro] Review announcement > > > > > Haven't done this in a long time. But who says neurons are noisy? > > > > From the point of view of information theory, why isn't the apperance > of > noise expected in a highly optimized coding scheme? And why isn't > synchrony > to be avoided as redundency. Engineers avoid it, why shouldn't evolution. > > > > Just curious. > > > > Jim bower > > Sent via BlackBerry by AT&T > > > > -----Original Message----- > > From: Nathan Urban > > > > Date: Fri, 18 Jul 2008 08:41:22 > > To: > > Subject: [Comp-neuro] Review announcement > > > > > > Review announcement > > > > This review describes a constructive role for noise in synchronizing > > populations of neurons and should be of interest to computaional > > neurosciuentists. > > > > > > Trends Neurosci. 2008 Jul 4. [Epub ahead of print] > > > Reliability, synchrony and noise. > > Ermentrout GB, Gal?n RF, Urban NN. > > > > The brain is noisy. Neurons receive tens of thousands of highly > > fluctuating inputs and generate spike trains that appear highly > > irregular. Much of this activity is spontaneous - uncoupled to overt > > stimuli or motor outputs - leading to questions about the functional > > impact of this noise. Although noise is most often thought of as > > disrupting patterned activity and interfering with the encoding of > > stimuli, recent theoretical and experimental work has shown that noise > > can play a constructive role - leading to increased reliability or > > regularity of neuronal firing in single neurons and across populations. > > These results raise fundamental questions about how noise can influence > > neural function and computation. > > > > PMID: 18603311 [PubMed - as supplied by > publisher] > > > > > http://www.sciencedirect.com/science?_ob=ArticleURL&_udi=B6T0V-4SXC918-1&_us > er=525223&_rdoc=1&_fmt=&_orig=search&_sort=d&view=c&_acct=C000026389&_versio > n=1&_urlVersion=0&_userid=525223&md5=22a86291fe13cd59541d841f692f24a2 > > _______________________________________________ > > Comp-neuro mailing list > > Comp-neuro@neuroinf.org > > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > > > ---------------------------------------------------------------------------- > ---- > > > > _______________________________________________ > > Comp-neuro mailing list > > Comp-neuro@neuroinf.org > > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > > _______________________________________________ > Comp-neuro mailing > list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > -- "In God We trust, Everyone else needs to bring data to the table" -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2a66a9d7/attachment-0001.html From bower at uthscsa.edu Tue Jul 22 16:17:57 2008 From: bower at uthscsa.edu (jim bower) Date: Tue Jul 22 16:56:44 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> Message-ID: <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> I am actually in a remote part of brazil at the moment, so limited to typing on my blackberry. However, yes I was curious if a discussion could be induced. That was originally what this mailing list was set up for, I know, because I started it. ;-). However things have become a bit complacent so I figured what the heck. Again limited in my ability to respond but a couple of things. I think as computational neurobiologists or scientists in general, we need to be aware of the extent which what we can measure (oscillations, synchronous spikes, etc) limits the way we think about how things work. Many many years ago now when cortical oscillations became more generally interesting to people once found in visual cortex we suggested based on our realistic cortical models that they were an epiphenomina more (loosly) reflecting and underlying mechanism for coordinating communication and processing between regions than carriers of any information themselves. I continue to believe or set my primary assumption that until proven otherwise, every spike is significant for something and worse yet so is the lack of a spike. (Certainly in digital coding 0s are as important as 1s. Yes "serious scientists" prefer more constrained and defined discussions than this. - but we can easily get lost "drinking our own whisky". As a famous computational math-bio guy is fond of saying. ;-) Truth is all these issues really remain wide open. But and the big but, no evidence that nature is sloppy or unsophisticated. One last point, the assumption that in fact nature is very sophisticated and that the structure of the brain deeply reflects a complex, sophisticated function pushes in the direction of first building models reflecting that structure, even if you are still clueless about function. I am in brazil teaching at the latin american school for computational neuroscience, where realistic modeling lives on. ;-) Best to all Jim Sent via BlackBerry by AT&T -----Original Message----- From: "Sven Abrahamse" Date: Tue, 22 Jul 2008 15:49:56 To: Cc: Subject: Re: [Comp-neuro] Review announcement _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From m.kaiser at newcastle.ac.uk Tue Jul 22 16:33:41 2008 From: m.kaiser at newcastle.ac.uk (Marcus Kaiser) Date: Tue Jul 22 16:57:35 2008 Subject: [Comp-neuro] Re: Review announcement In-Reply-To: References: Message-ID: All, >Thus, noise from molecular sources can have an impact in the highly >energy efficient and thus compact circuits >of our brains, setting lower limits to how precise we can encode >information. How to design (like an engineer) a brain's circuits is >thus a multi-factor trade-off problem...and to me quite fascinating. The role of noise, in the form of inaccurate processing results, was (first?) discussed by John von Neumann (The Computer and the Brain, Yale Univ Press, 1958). He was arguing that noise within a processing chain could change the final result of the calculation through error propagation (even more so for nonlinear systems). One way to reduce such an effect of noise is reducing the number of processing steps. Indeed, neural networks seem to be optimized for short processing paths (http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.00200 95 ). Changing the network structure (topology) to reduce the negative effects of noise seems to be more energy-efficient than active mechanisms (e.g. filtering or inhibition). Marcus -- Marcus Kaiser, Ph.D. School of Computing Science Newcastle University Claremont Tower Newcastle upon Tyne NE1 7RU, U.K. Phone: +44 191 222 8161 Fax: +44 191 222 8232 http://www.biological-networks.org/ From vcu at cs.stir.ac.uk Tue Jul 22 16:35:42 2008 From: vcu at cs.stir.ac.uk (Vassilis Cutsuridis) Date: Tue Jul 22 17:26:05 2008 Subject: [Comp-neuro] CfP Journal Special Issue on Neural Models of Cortical Microcircuits Message-ID: <49BD30C2416B46C8AA67EC589C8C0428@cs.ad.stir.ac.uk> First Call for Papers: Journal Special Issue on == Neural Models of Cortical Microcircuits == Guest Editors: J.G. Taylor, T. Wennekers, B.P. Graham, I. Vida, V. Cutsuridis Special issue of the Elsevier Journal of Neural Networks http://helen.pion.ac.uk/~thomas/microcircuits08 = SCOPE = To understand how perception, attention, action, learning and memory work, we need to gather data from multiple levels of complexity and from various brain states (normal and diseased) and integrate them at the brain-scale level. We need to identify the neuronal groups involved in these functions, their laminar distributions and their different types of neurons, draw detailed circuit diagrams, determine the forms of synaptic transmission and plasticity between different neurons and study the dynamics of the cortical microcircuits at the cellular and synaptic level that comprise these neuronal groups. Recent years have witnessed a dramatic accumulation of knowledge about the morphological, physiological and molecular characteristics, as well as the connectivity and synaptic properties of cortical neurons. Despite these advances, however, only limited insight was gained into the computational function of the neurons; in particular, the role of the various types of interneurons remains elusive. Mathematical and computational microcircuit models play an instrumental role in exploring microcircuit functions and facilitate the dissection the operations performed by diverse interneurons. The goal of the special issue is to provide a snapshot and a resum? of the current state-of-the-art of the ongoing research avenues concerning cortical microcircuits with particular emphasis on the functional roles of the various inhibitory interneurons in information processing within normal and diseased behavioural and cognitive states. The emphasis will be on computational models that are tightly grounded on experimental data = SPECIFIC AIMS = - The interaction between the local micro circuit activity and global processing to achieve the desired overall processing functionality observed, say in perception and action, attention, learning and memory - Microcircuit architectures: networks of principal and inhibitory inter-neurons within and between lamina, columns, mini-columns, modules, areas and/or across areas in the brain and their functional roles in the network. -- Neo-cortex -- Hippocampus -- Sensory and Motor Systems - Cross-comparison of architectures from different brain areas - Identified computations performed by each type of neuron in a network - Identified modes of operation of a neuronal type and how they are related potentially to behaviour and cognition - What synaptic plasticity rules are used = SUBMISSIONS = Deadline for submissions is December 1st, 2008. Electronic submissions for the Neural Networks journal can be found under http://ees.elsevier.com/neunet/ Please indicate in your cover letter that your article is for the special issue "Neural Models of Cortical Microcircuits" Regards, Vassilis ---------------------------------------------------------------------------------- Dr. Vassilis Cutsuridis Department of Computing Science and Mathematics University of Stirling Stirling FK9 4LA SCOTLAND Tel: +44 1786 467422 Fax: +44 1786 464551 Email: vcu@cs.stir.ac.uk Web: http://www.cs.stir.ac.uk/~vcu/ -- Academic Excellence at the Heart of Scotland. The University of Stirling is a charity registered in Scotland, number SC 011159. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/149040a5/attachment-0001.html From vcu at cs.stir.ac.uk Tue Jul 22 17:37:05 2008 From: vcu at cs.stir.ac.uk (Vassilis Cutsuridis) Date: Tue Jul 22 18:10:50 2008 Subject: [Comp-neuro] CfP Journal Special Issue on Neural Models of Cortical Microcircuits (correction) Message-ID: <530937473F2B49779E4317EF01988D18@cs.ad.stir.ac.uk> First Call for Papers: Journal Special Issue on == Neural Models of Cortical Microcircuits == Guest Editors: J.G. Taylor, T. Wennekers, B.P. Graham, I. Vida, V. Cutsuridis Special issue of the Elsevier Journal of Neural Networks http://helen.pion.ac.uk/microcircuits08 = SCOPE = To understand how perception, attention, action, learning and memory work, we need to gather data from multiple levels of complexity and from various brain states (normal and diseased) and integrate them at the brain-scale level. We need to identify the neuronal groups involved in these functions, their laminar distributions and their different types of neurons, draw detailed circuit diagrams, determine the forms of synaptic transmission and plasticity between different neurons and study the dynamics of the cortical microcircuits at the cellular and synaptic level that comprise these neuronal groups. Recent years have witnessed a dramatic accumulation of knowledge about the morphological, physiological and molecular characteristics, as well as the connectivity and synaptic properties of cortical neurons. Despite these advances, however, only limited insight was gained into the computational function of the neurons; in particular, the role of the various types of interneurons remains elusive. Mathematical and computational microcircuit models play an instrumental role in exploring microcircuit functions and facilitate the dissection the operations performed by diverse interneurons. The goal of the special issue is to provide a snapshot and a resum? of the current state-of-the-art of the ongoing research avenues concerning cortical microcircuits with particular emphasis on the functional roles of the various inhibitory interneurons in information processing within normal and diseased behavioural and cognitive states. The emphasis will be on computational models that are tightly grounded on experimental data = SPECIFIC AIMS = - The interaction between the local micro circuit activity and global processing to achieve the desired overall processing functionality observed, say in perception and action, attention, learning and memory - Microcircuit architectures: networks of principal and inhibitory inter-neurons within and between lamina, columns, mini-columns, modules, areas and/or across areas in the brain and their functional roles in the network. -- Neo-cortex -- Hippocampus -- Sensory and Motor Systems - Cross-comparison of architectures from different brain areas - Identified computations performed by each type of neuron in a network - Identified modes of operation of a neuronal type and how they are related potentially to behaviour and cognition - What synaptic plasticity rules are used = SUBMISSIONS = Deadline for submissions is December 1st, 2008. Electronic submissions for the Neural Networks journal can be found under http://ees.elsevier.com/neunet/ Please indicate in your cover letter that your article is for the special issue "Neural Models of Cortical Microcircuits" Regards, Vassilis ---------------------------------------------------------------------------------- Dr. Vassilis Cutsuridis Department of Computing Science and Mathematics University of Stirling Stirling FK9 4LA SCOTLAND Tel: +44 1786 467422 Fax: +44 1786 464551 Email: vcu@cs.stir.ac.uk Web: http://www.cs.stir.ac.uk/~vcu/ -- Academic Excellence at the Heart of Scotland. The University of Stirling is a charity registered in Scotland, number SC 011159. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/b211317a/attachment.html From Etienne.Roesch at pse.unige.ch Tue Jul 22 17:28:04 2008 From: Etienne.Roesch at pse.unige.ch (Etienne B. Roesch) Date: Tue Jul 22 18:47:04 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> <318893.22356.qm@web65511.mail.ac4.yahoo.com> <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> Message-ID: <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Yeah, I am loving the discussion! More, more! As an early postdoc, I still have in my working memory the classes I went through in grad school, and I remember this connectionist lecturer arguing that noise was actually a good thing for classifier- like systems (and by extension neural nets, and by extension plausible neural nets -- which are not classifiers stricto senso I agree) in that it allows an easier discrimination of the input in a probabilistic context. Given that redundancy of information/signal plays a big part in how the brain does the job, wouldn't noise be a clever mechanism to discriminate close-to-threshold stimuli? What do you think? Best regards, Le 22 juil. 08 ? 17:17, jim bower a ?crit : > I am actually in a remote part of brazil at the moment, so limited > to typing on my blackberry. Impressive typing skills, I have to admit. ;-) > However, yes I was curious if a discussion could be induced. That > was originally what this mailing list was set up for, I know, > because I started it. ;-). However things have become a bit > complacent so I figured what the heck. > > Again limited in my ability to respond but a couple of things. I > think as computational neurobiologists or scientists in general, we > need to be aware of the extent which what we can measure > (oscillations, synchronous spikes, etc) limits the way we think > about how things work. Many many years ago now when cortical > oscillations became more generally interesting to people once found > in visual cortex we suggested based on our realistic cortical > models that they were an epiphenomina more (loosly) reflecting and > underlying mechanism for coordinating communication and processing > between regions than carriers of any information themselves. I > continue to believe or set my primary assumption that until proven > otherwise, every spike is significant for something and worse yet > so is the lack of a spike. > (Certainly in digital coding 0s are as important as 1s. > > Yes "serious scientists" prefer more constrained and defined > discussions than this. - but we can easily get lost "drinking our > own whisky". As a famous computational math-bio guy is fond of saying. > > ;-) > > Truth is all these issues really remain wide open. > > But and the big but, no evidence that nature is sloppy or > unsophisticated. > > One last point, the assumption that in fact nature is very > sophisticated and that the structure of the brain deeply reflects a > complex, sophisticated function pushes in the direction of first > building models reflecting that structure, even if you are still > clueless about function. > > I am in brazil teaching at the latin american school for > computational neuroscience, where realistic modeling lives on. ;-) > > Best to all > > Jim > > Sent via BlackBerry by AT&T ----- Etienne Roesch Department of Computing Imperial College London SW7 2AZ -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/7b4ca486/attachment-0001.html From bower at uthscsa.edu Tue Jul 22 19:26:48 2008 From: bower at uthscsa.edu (jim bower) Date: Wed Jul 23 11:06:03 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com><280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry><39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: <525972290-1216747672-cardhu_decombobulator_blackberry.rim.net-1640555375-@bxe111.bisx.prod.on.blackberry> What is the evidence the brain uses redundency? How do you refine redundency? Is a 1024 x 1024 screen redundent because some pixals sometimes carry the same image (depending on the image being portrayed). And what are the consequences for all of our ideas about the brain that most of the neuronal data we have is in response to impoverished stimuli in anesthetized animals who otherwise are mostly bred in captivity and themselves live impoverished lives? We have never built anything near the complexity or capabilty of any brain I know, how can our understanding of the systems we can engineer have any hope of revealing the nature of the neural code. By the way if cryptographers were as fast to declare the signals they look at as statistical noise, they would be out of a job. Jim Sent via BlackBerry by AT&T -----Original Message----- From: "Etienne B. Roesch" (UNIGE) Date: Tue, 22 Jul 2008 18:28:04 To: Cc: Subject: Re: [Comp-neuro] Review announcement _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From malcolmdean at gmail.com Tue Jul 22 23:16:34 2008 From: malcolmdean at gmail.com (Malcolm Dean) Date: Wed Jul 23 11:06:08 2008 Subject: [Comp-neuro] Noise Message-ID: <417b04640807221416x534f9e70vcb860758809cc784@mail.gmail.com> Skipped content of type multipart/alternative-------------- next part -------------- A non-text attachment was scrubbed... Name: science(21) Type: application/octet-stream Size: 1070 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/science21-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: science(22) Type: application/octet-stream Size: 1470 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/science22-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: 367270a0.ris Type: application/octet-stream Size: 451 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/367270a0-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: citmgr(4) Type: application/octet-stream Size: 1015 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/citmgr4-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: 10.1371%2Fjournal.pcbi.0030079.ris Type: application/octet-stream Size: 2057 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/10.13712Fjournal.pcbi.0030079-0001.obj -------------- next part -------------- A non-text attachment was scrubbed... Name: 10.1371%2Fjournal.pcbi.0020095.ris Type: application/octet-stream Size: 2077 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2e3efa00/10.13712Fjournal.pcbi.0020095-0001.obj From bard at math.pitt.edu Tue Jul 22 21:22:43 2008 From: bard at math.pitt.edu (G. Bard Ermentrout) Date: Wed Jul 23 11:07:13 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> <318893.22356.qm@web65511.mail.ac4.yahoo.com> <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: For what it is worth - I have puzzled over the ubiquity of oscillations in the CNS and still wonder what they are good for. Jim and others argue epiphenomena, and this could still be correct, but it is real hard for me to believe that nature would ignore a free byproduct like this. One thing about oscillations is that they have associated with them a zero eigenvalue at the single cell, microcircuit or other level and what this does is it makes it very eay to modulate the timing of their spikes. Much more so than with fixed points. Thus it very easy from the point of view of efficiency to move the spikes around in sych a way as to e.g. compute correlations via the stochastic synchrony mechanism and thus propagate feedfoward synchronous or correlated activity to other areas or layers. Synchrony or near synchrony is very efficient at propagating in feedforward networks. Oscillations make it real easy to read out correlations and also make it very easy to quickly desynchronize groups with simple modulation of their intrinsic dynamaics - e.g. ACh which can greatly affect how a neuron responds to the timing of an input and to other neurons to which it is attached. >From Rome with one vino too many, bard Ermentrout From minai_ali at yahoo.com Tue Jul 22 19:41:48 2008 From: minai_ali at yahoo.com (Ali Minai) Date: Wed Jul 23 11:07:29 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: <989871.36961.qm@web65515.mail.ac4.yahoo.com> Noise does this and much, much more. It can inject variety, break symmetry, generate novelty, provide energy, facilitate search, carry signal, and do many other things. Indeed, the only time noise is really a problem is when one is trying to do achieve a pre-determined goal (e.g., following a pre-computed trajectory). Since natural systems - notably the nervous system - rarely (if ever) try to do this, they thrive on noise. Perhaps we should give the phenomenon a less pejorative name. "Noise" signals such a linear mindset:-). Ali --------------------------------------------------------------------- Ali A. Minai Associate Professor Associate Head for Electrical Engineering Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: aminai@ececs.uc.edu ????????? minai_ali@yahoo.com WWW: http://www.ececs.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Tue, 7/22/08, Etienne B. Roesch wrote: From: Etienne B. Roesch Subject: Re: [Comp-neuro] Review announcement To: comp-neuro@neuroinf.org Cc: comp-neuro-bounces@neuroinf.org Date: Tuesday, July 22, 2008, 11:28 AM Yeah, I am loving the discussion! More, more! As an early postdoc, I still have in my working memory the classes I went through in grad school, and I remember this connectionist lecturer arguing that noise was actually a good thing for classifier-like systems (and by extension neural nets, and by extension plausible neural nets -- which are not classifiers stricto senso I agree) in that it allows an easier discrimination of the input in a probabilistic context. Given that redundancy of information/signal plays a big part in how the brain does the job, wouldn't noise be a clever mechanism to discriminate close-to-threshold stimuli? What do you think? Best regards, Le 22 juil. 08 ? 17:17, jim bower a ?crit : I am actually in a remote part of brazil at the moment, so limited to typing on my blackberry. Impressive typing skills, I have to admit. ;-) However, yes I was curious if a discussion could be induced. That was originally what this mailing list was set up for, I know, because I started it. ;-). However things have become a bit complacent so I figured what the heck. Again limited in my ability to respond but a couple of things. I think as computational neurobiologists or scientists in general, we need to be aware of the extent which what we can measure (oscillations, synchronous spikes, etc) limits the way we think about how things work. Many many years ago now when cortical oscillations became more generally interesting to people once found in visual cortex we suggested based on our realistic cortical models that they were an epiphenomina more (loosly) reflecting and underlying mechanism for coordinating communication and processing between regions than carriers of any information themselves. I continue to believe or set my primary assumption that until proven otherwise, every spike is significant for something and worse yet so is the lack of a spike. (Certainly in digital coding 0s are as important as 1s. Yes "serious scientists" prefer more constrained and defined discussions than this. - but we can easily get lost "drinking our own whisky". As a famous computational math-bio guy is fond of saying. ;-) Truth is all these issues really remain wide open. But and the big but, no evidence that nature is sloppy or unsophisticated. One last point, the assumption that in fact nature is very sophisticated and that the structure of the brain deeply reflects a complex, sophisticated function pushes in the direction of first building models reflecting that structure, even if you are still clueless about function. I am in brazil teaching at the latin american school for computational neuroscience, where realistic modeling lives on. ;-) Best to all Jim Sent via BlackBerry by AT&T -----Etienne RoeschDepartment of ComputingImperial CollegeLondon SW7 2AZ_______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080722/2b5fe52d/attachment.html From bower at uthscsa.edu Wed Jul 23 11:50:48 2008 From: bower at uthscsa.edu (jim bower) Date: Wed Jul 23 12:08:11 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com><280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry><39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: <196425576-1216806710-cardhu_decombobulator_blackberry.rim.net-1010966473-@bxe111.bisx.prod.on.blackberry> I see your vino and raise you two caipirinhas. ;-) "Ease of propegating synchronous signals". Doesn't this fundamentally depend on neurons essentially being temporal / spatial summing devices? Which they aren't at least not the ones I deal with. BTW when I suggest that oscillations are epiphenomina, I mean when recorded at the level of extracellular field potentials. I have no doubt myself that the brain and neurons care about periodicity. This is clear as has been pointed out here already in the periodic behavior of motor systems and therefore the neurons that coordinate movement (CPGs). My personal suspicions revolve around the specific significance of synchrony, which it seems to me is inevitably tied to an integrate and fire assumption about neuronal processing (and core therefore to most abstract models of neurons and networks). With respect to "noise" again, I personally thing the term should be banned or barring that, we should agree on a common definition which is surprisingly hard to do even in computational neuroscience. Finally, the most tightly regulated (I.e. Neuronal event associated with the largest number and probably most complex ion channels) is spike initiation and in particular, the regulation of spike to spike firing patterns. Doesn't that suggest that the timing of individual spikes and spike trains is critical to signal transduction? This is what raises concern for me when I am told that neurons are intrinsically noisy devices. They sure spend a lot of energy controlling the timing of their outputs. Jim Sent via BlackBerry by AT&T -----Original Message----- From: "G. Bard Ermentrout" Date: Tue, 22 Jul 2008 15:22:43 Cc: ; Subject: Re: [Comp-neuro] Review announcement For what it is worth - I have puzzled over the ubiquity of oscillations in the CNS and still wonder what they are good for. Jim and others argue epiphenomena, and this could still be correct, but it is real hard for me to believe that nature would ignore a free byproduct like this. One thing about oscillations is that they have associated with them a zero eigenvalue at the single cell, microcircuit or other level and what this does is it makes it very eay to modulate the timing of their spikes. Much more so than with fixed points. Thus it very easy from the point of view of efficiency to move the spikes around in sych a way as to e.g. compute correlations via the stochastic synchrony mechanism and thus propagate feedfoward synchronous or correlated activity to other areas or layers. Synchrony or near synchrony is very efficient at propagating in feedforward networks. Oscillations make it real easy to read out correlations and also make it very easy to quickly desynchronize groups with simple modulation of their intrinsic dynamaics - e.g. ACh which can greatly affect how a neuron responds to the timing of an input and to other neurons to which it is attached. >From Rome with one vino too many, bard Ermentrout _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From harry.erwin at sunderland.ac.uk Wed Jul 23 12:47:49 2008 From: harry.erwin at sunderland.ac.uk (Harry Erwin) Date: Wed Jul 23 13:24:31 2008 Subject: [Comp-neuro] Oscillations In-Reply-To: References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> <318893.22356.qm@web65511.mail.ac4.yahoo.com> <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: An issue I've been thinking about recently is the evidence that goal- oriented plans can be replayed at various speeds and in both forward and time-reversed directions. In goal-directed behaviour of bats, the animal first plans ahead to the target capture (or perhaps retrieves an appropriate plan from memory). Then later during the capture process, if the target turns out to be inedible, the bat will sheer off as late as 30 msec prior to contact. I guestimate that the original capture plan was generated or retrieved in about 5% of the time necessary to execute it, and the back propagation through time of revised reward estimates takes place in about the same time. I suspect the plan is represented as a set of discrete intermediate subgoals, and that there is an oscillatory process that steps through the subgoals to replay the plan. Chip Levy's evidence about sharp waves during sleep suggests some mechanisms that would allow the speed and direction of the process to be varied. -- "an academic who listens to pleas of convenience before publishing his research risks calling into doubt the whole of his determination to find the truth." (Russell 1993) Harry Erwin From ucganlb at ucl.ac.uk Wed Jul 23 13:10:59 2008 From: ucganlb at ucl.ac.uk (Neil Burgess) Date: Wed Jul 23 13:44:03 2008 Subject: [Comp-neuro] oscillations can be useful Message-ID: <200807231134.m6NBYICT011225@mailer04.ua.ac.be> One area where oscillations seem to provide useful information is the hippocampal theta rhythm, against which the phase of firing of place cells1 and grid cells2 indicate the distance travelled by a rat through the firing field(s) of the cell. It seems likely that this phenomenon, and the regular repeated spatial firing pattern of grid cells, results from the interference of oscillatory influences on these cells' membrane potentials1345. Best wishes, Neil 1. O'Keefe and Recce 1993 Hippocampus 3, 317-30 2. Hafting et al 2008 Nature 453, 1248-52. 3. Lengyel et al 2003 Hippocampus 13, 700-14. 4. Burgess et al 2007 Hippocampus 17, 801-12. 5. Giocomo et al 2007 Science 315, 1719-22. Neil Burgess Institute of Cognitive Neuroscience University College London n.burgess@ucl.ac.uk -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080723/cd5181ff/attachment.html From lyle at biomedicale.univ-paris5.fr Wed Jul 23 12:51:04 2008 From: lyle at biomedicale.univ-paris5.fr (Lyle Graham) Date: Wed Jul 23 13:44:48 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <196425576-1216806710-cardhu_decombobulator_blackberry.rim.net-1010966473-@bxe111.bisx.prod.on.blackberry> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com><280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry><39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> <196425576-1216806710-cardhu_decombobulator_blackberry.rim.net-1010966473-@bxe111.bisx.prod.on.blackberry> Message-ID: <48870D18.7080400@biomedicale.univ-paris5.fr> Bonjour, un Pastis ici Nothing particularly new, but one way to frame it is: A recorded signal from [insert favorite system, e.g. impoverished anesthetized cat visual cortex] during the presentation of some defined stimulus has three interacting components - that which is evoked by the stimulus, that which reflects the physics of the structure, and that arising from the code of whatever other functional processing, broadly defined, is going on at the moment. The stochastic part of the second component corresponds to noise in the engineering sense. When there are only weak assumptions as to the qualitative nature of the third component, it is often useful to quantify it with the same framework as the stochastic second component. This parsing leads to questions such as, given everything else the animal is thinking about right now, how much can this particular signal describe this particular stimulus? Also, it reminds me that "noise" is a function of the observer - thus in a given case if there is no useful information in the echo from other processing, that motivates the question how a stochastic component can be a functionally bug and/or feature. n.b. I would be willing to bet that for a large body of data, the "other processing going on" is as rich and significant - thus indistinguishable - whether the beast grew up in the savannah or the animal facility. Which means that there are plenty of ideas still to be gleaned from impoverished and even anesthetized animals. Lyle -- Lyle J. Graham Laboratory of Neurophysics and Physiology, CNRS UMR 8119 www.neurophys.biomedicale.univ-paris5.fr/~graham Universit? Paris Descartes 45 rue des Saint-P?res, 75006 Paris Tel: 33 1 42 86 20 92 Fax: 33 1 49 27 90 62 Secr?tariat: 33 1 42 86 21 38 jim bower wrote: > I see your vino and raise you two caipirinhas. ;-) > > > "Ease of propegating synchronous signals". Doesn't this fundamentally depend on neurons essentially being temporal / spatial summing devices? Which they aren't at least not the ones I deal with. > > BTW when I suggest that oscillations are epiphenomina, I mean when recorded at the level of extracellular field potentials. I have no doubt myself that the brain and neurons care about periodicity. This is clear as has been pointed out here already in the periodic behavior of motor systems and therefore the neurons that coordinate movement (CPGs). > > My personal suspicions revolve around the specific significance of synchrony, which it seems to me is inevitably tied to an integrate and fire assumption about neuronal processing (and core therefore to most abstract models of neurons and networks). > > With respect to "noise" again, I personally thing the term should be banned or barring that, we should agree on a common definition which is surprisingly hard to do even in computational neuroscience. > > Finally, the most tightly regulated (I.e. Neuronal event associated with the largest number and probably most complex ion channels) is spike initiation and in particular, the regulation of spike to spike firing patterns. Doesn't that suggest that the timing of individual spikes and spike trains is critical to signal transduction? This is what raises concern for me when I am told that neurons are intrinsically noisy devices. They sure spend a lot of energy controlling the timing of their outputs. > > Jim > Sent via BlackBerry by AT&T > > -----Original Message----- > From: "G. Bard Ermentrout" > > Date: Tue, 22 Jul 2008 15:22:43 > Cc: ; > Subject: Re: [Comp-neuro] Review announcement > > > For what it is worth - I have puzzled over the ubiquity of oscillations in > the CNS and still wonder what they are good for. Jim and others argue > epiphenomena, and this could still be correct, but it is real hard for me > to believe that nature would ignore a free byproduct like this. One thing > about oscillations is that they have associated with them a zero > eigenvalue at the single cell, microcircuit or other level and what this > does is it makes it very eay to modulate the timing of their spikes. Much > more so than with fixed points. Thus it very easy from the point of view > of efficiency to move the spikes around in sych a way as to e.g. compute > correlations via the stochastic synchrony mechanism and thus propagate > feedfoward synchronous or correlated activity to other areas or layers. > Synchrony or near synchrony is very efficient at propagating in > feedforward networks. Oscillations make it real easy to read out > correlations and also make it very easy to quickly desynchronize groups > with simple modulation of their intrinsic dynamaics - e.g. ACh which can > greatly affect how a neuron responds to the timing of an input and to > other neurons to which it is attached. > > >From Rome with one vino too many, > > bard Ermentrout > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > ------------------------------------------------------------------------ > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > jim bower wrote: > What is the evidence the brain uses redundency? How do you refine redundency? Is a 1024 x 1024 screen redundent because some pixals sometimes carry the same image (depending on the image being portrayed). And what are the consequences for all of our ideas about the brain that most of the neuronal data we have is in response to impoverished stimuli in anesthetized animals who otherwise are mostly bred in captivity and themselves live impoverished lives? > > We have never built anything near the complexity or capabilty of any brain I know, how can our understanding of the systems we can engineer have any hope of revealing the nature of the neural code. > > By the way if cryptographers were as fast to declare the signals they look at as statistical noise, they would be out of a job. > > > Jim > Sent via BlackBerry by AT&T -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080723/039de420/attachment.html From minai_ali at yahoo.com Wed Jul 23 14:52:24 2008 From: minai_ali at yahoo.com (Ali Minai) Date: Wed Jul 23 15:12:11 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: Message-ID: <676130.90790.qm@web65515.mail.ac4.yahoo.com> I would add to this the ideas recently advanced by Izhikevich and Hoppensteadt (among others) that subthreshold oscillations can effectively multiplex signals so that different sub-populations of neurons only respond to spikes that occur near the peaks of their oscillations. Of course, one still has to explain how various sub-populations come to have their own synchronized sub-threshold oscillations in the first place, but it is an interesting mechanism. I certainly agree with Bard Ermentrout that neural oscillations are too ubiquitous a phenomenon to be left unexploited by evolution. Neil Burgess already pointed out one potential function in the hippocampus. Also, what about a simple sequencing function? Walter Freeman talks about a neural "shutter", and some sort of "clocking" is implicit in many models of cognitive function. Ali --------------------------------------------------------------------- Ali A. Minai Associate Professor Associate Head for Electrical Engineering Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: aminai@ececs.uc.edu ????????? minai_ali@yahoo.com WWW: http://www.ececs.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Tue, 7/22/08, G. Bard Ermentrout wrote: From: G. Bard Ermentrout Subject: Re: [Comp-neuro] Review announcement To: Cc: comp-neuro@neuroinf.org, comp-neuro-bounces@neuroinf.org Date: Tuesday, July 22, 2008, 3:22 PM For what it is worth - I have puzzled over the ubiquity of oscillations in the CNS and still wonder what they are good for. Jim and others argue epiphenomena, and this could still be correct, but it is real hard for me to believe that nature would ignore a free byproduct like this. One thing about oscillations is that they have associated with them a zero eigenvalue at the single cell, microcircuit or other level and what this does is it makes it very eay to modulate the timing of their spikes. Much more so than with fixed points. Thus it very easy from the point of view of efficiency to move the spikes around in sych a way as to e.g. compute correlations via the stochastic synchrony mechanism and thus propagate feedfoward synchronous or correlated activity to other areas or layers. Synchrony or near synchrony is very efficient at propagating in feedforward networks. Oscillations make it real easy to read out correlations and also make it very easy to quickly desynchronize groups with simple modulation of their intrinsic dynamaics - e.g. ACh which can greatly affect how a neuron responds to the timing of an input and to other neurons to which it is attached. >From Rome with one vino too many, bard Ermentrout _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080723/e0bf3f74/attachment-0001.html From bower at uthscsa.edu Wed Jul 23 16:12:53 2008 From: bower at uthscsa.edu (jim bower) Date: Wed Jul 23 16:29:08 2008 Subject: [Comp-neuro] Oscillations In-Reply-To: References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com><280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry><39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> Message-ID: <669309838-1216822435-cardhu_decombobulator_blackberry.rim.net-947387123-@bxe111.bisx.prod.on.blackberry> Many interestimg ideas in this discussion, which we can probably anticipate will continue in person at the upcoming CNS meeting, which was the initial origin of this mailing list -- so perhaps this is an interesting way to prime the pump for that meeting. I would caution however about what I have come to think of as "the tyrany of ideas" in computational neuroscience or biology in general. The brain is probably about the most dangerous place you can think of to look for proof of largely a priori ideas. There are many examples, the Marr/Albus model of cerebellar learning for example has produced almost 50 years of efforts by experimentalists and theorist/modelers to generate evidence it is true. Much of the work on oscillations is similar, an idea (from machine vision and AI originally related to the imagined problem of segmentation) in search of evidence. I suppose it is obvious, but my overall point here is that we have to be very very careful about our assumptions given our ignorence. (And by the way sorry for the spelling and other errors - writing from my blackberry on a beach in Brazil - so many opportunities for errors. - context being everything or nearly everything. Jim Sent via BlackBerry by AT&T -----Original Message----- From: Harry Erwin Date: Wed, 23 Jul 2008 11:47:49 To: Subject: [Comp-neuro] Oscillations An issue I've been thinking about recently is the evidence that goal- oriented plans can be replayed at various speeds and in both forward and time-reversed directions. In goal-directed behaviour of bats, the animal first plans ahead to the target capture (or perhaps retrieves an appropriate plan from memory). Then later during the capture process, if the target turns out to be inedible, the bat will sheer off as late as 30 msec prior to contact. I guestimate that the original capture plan was generated or retrieved in about 5% of the time necessary to execute it, and the back propagation through time of revised reward estimates takes place in about the same time. I suspect the plan is represented as a set of discrete intermediate subgoals, and that there is an oscillatory process that steps through the subgoals to replay the plan. Chip Levy's evidence about sharp waves during sleep suggests some mechanisms that would allow the speed and direction of the process to be varied. -- "an academic who listens to pleas of convenience before publishing his research risks calling into doubt the whole of his determination to find the truth." (Russell 1993) Harry Erwin _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From bower at uthscsa.edu Wed Jul 23 16:27:51 2008 From: bower at uthscsa.edu (jim bower) Date: Wed Jul 23 16:36:24 2008 Subject: [Comp-neuro] Noise, redundancy and biophysics In-Reply-To: References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry><318893.22356.qm@web65511.mail.ac4.yahoo.com><2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com><280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry><39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> <196425576-1216806710-cardhu_decombobulator_blackberry.rim.net-1010966473-@bxe111.bisx.prod.on.blackberry> Message-ID: <1496261028-1216823334-cardhu_decombobulator_blackberry.rim.net-1576141891-@bxe111.bisx.prod.on.blackberry> Sacha, I agree thus it would be nice to agree on a common definition - which has been very hard to do. A couple of questions: How can we possibly know what is "relevant" to a particular neuron? We can decide based on our experimental protocohls, but isn't that us imposing on "them". "the reliability and temporal precision with which a single presynaptic action potential results in a postsynaptic action potential can be low". In the cerebellum, the 150,000 excitatory inputs to each Purkinje cell don't appear to have any direct influence on the output of the cell - I suspect most excitatory synaptic inputs in the brain are actually doing exactly what they appear to be doing, nfluencing the local membrane. I believe that we neurobiologists may be far too "soma-centric" in thinking about how neuons and brains compute. Last isn't it intersting that the closer one gets to the periphery either on the sensory or the motor side the more "precise" the nervous system looks, yet somehow in the middle it looks like it needs to solve some signal to noise problem. But why isn't it simply loikely that we can better understand what is going on on either end, but in the middle we have no idea. Although I know bard and others who talk about noise are talking about it in a precise way -- however, I tend to lump the word "noise" like a number of other similar words in literature as a bin in which we put things we don't understand. I am reminded of Penzias (sorry about the spelling and wilson crawling around in their satalite dish with tin foil trying to remove what turned out to be (predicted) background cosmic radiation. (Please don't point out the role of theoretical modeling in realizing the truth. -- this modeling was done in the context of a science of simple things (physics) with among other things a common set of definitions). Jim Sent via BlackBerry by AT&T -----Original Message----- From: Sacha Nelson Date: Wed, 23 Jul 2008 09:40:56 To: Cc: Subject: Re: [Comp-neuro] Noise, redundancy and biophysics It is worth pointing out that "Signal," "noise," and "redundancy" are highly relative terms. From the perspective of recognizing the word, adding the missing "a" would be redundnt, while from the perspective of telling whether or not I can spell it is not. The decoding problem faced by neurons is really one of separating signals relevant to a particular computation from signals about other events not relevant to that computation, but these other signals are not necessarily noise in the classical (e.g. thermal) sense. In nearly every case in which it has been carefully looked at, neurons turn out to be exquisitely sensitive (e.g relative to behavior). Central synaptic connections, at least in the adult, turn out to be far more reliable than once thought. Much of the confusion about noise and redundancy can be traced to arguments based on systems (e.g. the neocortex of mammals) where the reliability and temporal precision with which a single presynaptic action potential results in a postsynaptic action potential can be low. But this is not a hardware limitation imposing intrinsic noise, it is a feature of the coding scheme, common to many central circuits, in which each neuron receives a very large number (e.g. ~10K) of individual inputs. The ability to precisely follow a single input is sacrificed in order to be able to detect correlation (and possibly higher moments) across multiple inputs. Precisely the opposite end of the spectrum is exemplified in many circuits closer to the sensory periphery (retinothalamic, auditory brainstem etc.) or motor output (neuromuscular junction). Here, hundreds of synaptic boutons comprise an individual connection and the temporal precision with which in input spike can give rise to an output spike can be measured in microseconds. Yet, the basic biophysical properties of the individual synapses and action potential encoding are essentially the same, or occupy the same range of properties, in both kinds of circuits. For the most part, in large brains at least, the circuit level seems fairly well insulated from "noise" at the molecular level (i.e. the stochasticity of channel openings). Of course there are some important exceptions, and this is less likely to hold in much smaller neurons in smaller circuits--e.g. worms which only have 300+ neurons and that lack action potentials entirely. -Sacha From levink at unimelb.edu.au Thu Jul 24 07:44:59 2008 From: levink at unimelb.edu.au (Levin Kuhlmann) Date: Thu Jul 24 10:13:26 2008 Subject: [Comp-neuro] 'NeuroEng 2008' Workshop in Melbourne: Call for Abstracts Message-ID: <9E6D018FA9D93F4FA29E3F612088A13802BBA488@IS-EX-BEV2.unimelb.edu.au> This is a call for abstracts for the 'NeuroEng 2008' Workshop to be held at the University of Melbourne in Melbourne Australia from Nov 20-22 2008 Abstracts are due by September 5. The flyer for the workshop and abstract submission templates can be downloaded from the webpage: http://www.neuroeng.unimelb.edu.au/NeuroEng2008.htm We hope to see you in November. Cheers -- Levin Kuhlmann Research Fellow Department of Electrical and Electronic Engineering University of Melbourne Parkville VIC 3010 Australia Ph: +613 8344 6689 Fax: +613 8344 6713 E-mail: l.kuhlmann@ee.unimelb.edu.au -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/881ac013/attachment.html From bard at math.pitt.edu Wed Jul 23 18:25:50 2008 From: bard at math.pitt.edu (G. Bard Ermentrout) Date: Thu Jul 24 10:22:44 2008 Subject: [Comp-neuro] Noise, redundancy and biophysics In-Reply-To: <1496261028-1216823334-cardhu_decombobulator_blackberry.rim.net-1576141891-@bxe111.bisx.prod.on.blackberry> References: <613297900-1216723647-cardhu_decombobulator_blackberry.rim.net-1121255964-@bxe111.bisx.prod.on.blackberry> <318893.22356.qm@web65511.mail.ac4.yahoo.com> <2f39a6f40807220649v68d93239y15fded5ff1aac9d6@mail.gmail.com> <280528948-1216736342-cardhu_decombobulator_blackberry.rim.net-1736439498-@bxe111.bisx.prod.on.blackberry> <39F6DD6E-B32D-44EA-9380-C530DC2A80C1@pse.unige.ch> <196425576-1216806710-cardhu_decombobulator_blackberry.rim.net-1010966473-@bxe111.bisx.prod.on.blackberry> <1496261028-1216823334-cardhu_decombobulator_blackberry.rim.net-1576141891-@bxe111.bisx.prod.on.blackberry> Message-ID: I want to once again emphasize that the "noise" that we discuss in our paper is not meant to be a distraction but rather is the signal in the sense that we do not know precisely the nature of the summed EPSPs and IPSPs that are felt at the site of AP generation. Thus, we are using a broadband input to study how a neuron responds (in some sense, like the ideas of many others, notably Bialek, Berger, et al) reliably when it is, in absence of input, firing an a fairly regular fashion. This is a somewhat restictive environment (admittedly) and we mean by noise the form of the signal - as it is repeatable, it is not noise in the engineering sense of something to be avoided. I agree with Sacha (at least at the hillock), that channel noise is not an issue. We were interested in whether a broadband signal could get through reliably in the presence of other signals that are of different origin (and treated as "noise" with respect to the first signal). Bard From rinkus at comcast.net Thu Jul 24 06:33:59 2008 From: rinkus at comcast.net (rod rinkus) Date: Thu Jul 24 10:22:48 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <989871.36961.qm@web65515.mail.ac4.yahoo.com> Message-ID: <000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> Excellent post, Dr. Minai, You may be interested in a poster that I just gave at AREADNE, which describes a new theory of how noise is used, as you suggest, to inject variety into the assignment of sparse distributed codes. Suppose you have a sparse distributed coding layer where N out of M cells are chosen for any representation. Suppose you have a mechanism for computing the familiarity, G (normalized between 0 and 1), of an input, X, before assigning a code to it. Finally, you have a mechanism that adds an amount of noise, inversely proportional to G (i.e., directly proportional to novelty) into the code selection process. For completely familiar inputs (G=1), the mechanism adds no noise, allowing the synaptic inputs to cells to be the dominant influence in choosing which cells become active, resulting in code completion (recognition). For completely novel inputs (G=0), the mechanism add so much noise that it swamps out the effects of synaptic inputs, making all cells equally likely to be chosen (i.e., N cells chosen from the uniform distribution). In this case, the expected intersection between the newly chosen code and any previously assigned code will be at the level of chance, resulting in code separation (learning). More generally, this mechanism varies the amount of noise added into the winner selection process from zero when G=1 to ?high enough to swamp out the synaptic inputs? when G=0. Overall, this mechanism ensures that the more similar the inputs are, the more similar (i.e., more overlapped) their codes are. The concept can be seen in more detail in my poster available at top of the publications page http://people.brandeis.edu/~grinkus/Publications of my Web site http://people.brandeis.edu/~grinkus/ Cheers Gerard Rinkus, PhD Visiting Scientist, Lisman Lab Volen Center for Complex Systems Brandeis University, Waltham, MA 781-736-3146 http://people.brandeis.edu/~grinkus/ _____ From: comp-neuro-bounces@neuroinf.org [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of Ali Minai Sent: Tuesday, July 22, 2008 1:42 PM To: comp-neuro@neuroinf.org; Etienne B. Roesch Subject: Re: [Comp-neuro] Review announcement Noise does this and much, much more. It can inject variety, break symmetry, generate novelty, provide energy, facilitate search, carry signal, and do many other things. Indeed, the only time noise is really a problem is when one is trying to do achieve a pre-determined goal (e.g., following a pre-computed trajectory). Since natural systems - notably the nervous system - rarely (if ever) try to do this, they thrive on noise. Perhaps we should give the phenomenon a less pejorative name. "Noise" signals such a linear mindset:-). Ali --------------------------------------------------------------------- Ali A. Minai Associate Professor Associate Head for Electrical Engineering Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: aminai@ececs.uc.edu minai_ali@yahoo.com WWW: http://www.ececs.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Tue, 7/22/08, Etienne B. Roesch wrote: From: Etienne B. Roesch Subject: Re: [Comp-neuro] Review announcement To: comp-neuro@neuroinf.org Cc: comp-neuro-bounces@neuroinf.org Date: Tuesday, July 22, 2008, 11:28 AM Yeah, I am loving the discussion! More, more! As an early postdoc, I still have in my working memory the classes I went through in grad school, and I remember this connectionist lecturer arguing that noise was actually a good thing for classifier-like systems (and by extension neural nets, and by extension plausible neural nets -- which are not classifiers stricto senso I agree) in that it allows an easier discrimination of the input in a probabilistic context. Given that redundancy of information/signal plays a big part in how the brain does the job, wouldn't noise be a clever mechanism to discriminate close-to-threshold stimuli? What do you think? Best regards, Le 22 juil. 08 ? 17:17, jim bower a ?crit : I am actually in a remote part of brazil at the moment, so limited to typing on my blackberry. Impressive typing skills, I have to admit. ;-) However, yes I was curious if a discussion could be induced. That was originally what this mailing list was set up for, I know, because I started it. ;-). However things have become a bit complacent so I figured what the heck. Again limited in my ability to respond but a couple of things. I think as computational neurobiologists or scientists in general, we need to be aware of the extent which what we can measure (oscillations, synchronous spikes, etc) limits the way we think about how things work. Many many years ago now when cortical oscillations became more generally interesting to people once found in visual cortex we suggested based on our realistic cortical models that they were an epiphenomina more (loosly) reflecting and underlying mechanism for coordinating communication and processing between regions than carriers of any information themselves. I continue to believe or set my primary assumption that until proven otherwise, every spike is significant for something and worse yet so is the lack of a spike. (Certainly in digital coding 0s are as important as 1s. Yes "serious scientists" prefer more constrained and defined discussions than this. - but we can easily get lost "drinking our own whisky". As a famous computational math-bio guy is fond of saying. ;-) Truth is all these issues really remain wide open. But and the big but, no evidence that nature is sloppy or unsophisticated. One last point, the assumption that in fact nature is very sophisticated and that the structure of the brain deeply reflects a complex, sophisticated function pushes in the direction of first building models reflecting that structure, even if you are still clueless about function. I am in brazil teaching at the latin american school for computational neuroscience, where realistic modeling lives on. ;-) Best to all Jim Sent via BlackBerry by AT&T ----- Etienne Roesch Department of Computing Imperial College London SW7 2AZ _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/7d026871/attachment-0001.html From reinoud at castafiore.cde.ua.ac.be Thu Jul 24 10:59:00 2008 From: reinoud at castafiore.cde.ua.ac.be (comp-neuro moderator) Date: Thu Jul 24 11:02:25 2008 Subject: [Fwd: Re: [Comp-neuro] Noise, redundancy and biophysics] Message-ID: <54451.143.169.8.71.1216889940.squirrel@castafiore.cde.ua.ac.be> ---------------------------- Original Message ---------------------------- Subject: Re: [Comp-neuro] Noise, redundancy and biophysics From: "jim bower" Date: Wed, July 23, 2008 7:33 pm To: bard@math.pitt.edu Cc: "Sacha Nelson" comp-neuro@neuroinf.org -------------------------------------------------------------------------- Great. This is then close, unless I am missing something, to the kind of analysis that Shannon would have attempted. Or am I missing something. Of course the problem is, once again, that "noisy" inputs (the sum of individual "noisy neurons") suffer from the same fundamental problem that we don't understand the code -- in fact let me move one step closer to the cliff with several other questions about the significance of action potentials: Who says that peak responses (as in tuning curves or PST hisograms or the like) represent the most significant responsiveness in a neuron. Or the stimUlous that generates the peak the most important stimlous - or the information that is being transmitted by the neuron to the next neurons? In biology its the steepest slope of a function where most of the action likely is. Second question, in signals represented across multiple channels, the significance of an event (action potential) is only in the context of the other channels. And as I suggested before the absence of an event might be as or more important than its presence. But absence at one moment might be more significant than at other moments. Now you have some waiting of significance for different things that didn't happen. Finally, theorists need to keep three things always in mind: 1) Experimentalists generally seek experimental stimuli and experimental conditions that produce the most obvious responses (biggest) responses in neurons 2) Experimentalists are generally story tellers and to this day seldom report the statistics around their results. 3) Experimentalists almost always show the best examples from their data, declaring those examples to be "representative" And 4) Most importantly, experimentalists seldom report all their results and in particular usually figure out ways to explain away contradictory results so as not to produce confusion. And the explaining away is done prepuublication so it never sees the light of day. This is a corellary to the "tyranny of ideas". It is the "tyranny of a clear story" and imposed by the story telling nature of most biology. Not to discourage anyone, this is what computational neuroscience has to change (and slowly is actually in my opinion). Jim ------Original Message------ From: G. Bard Ermentrout To: James Bower Cc: Sacha Nelson Cc: comp-neuro@neuroinf.org ReplyTo: bard@math.pitt.edu Sent: Jul 23, 2008 11:25 AM Subject: Re: [Comp-neuro] Noise, redundancy and biophysics I want to once again emphasize that the "noise" that we discuss in our paper is not meant to be a distraction but rather is the signal in the sense that we do not know precisely the nature of the summed EPSPs and IPSPs that are felt at the site of AP generation. Thus, we are using a broadband input to study how a neuron responds (in some sense, like the ideas of many others, notably Bialek, Berger, et al) reliably when it is, in absence of input, firing an a fairly regular fashion. This is a somewhat restictive environment (admittedly) and we mean by noise the form of the signal - as it is repeatable, it is not noise in the engineering sense of something to be avoided. I agree with Sacha (at least at the hillock), that channel noise is not an issue. We were interested in whether a broadband signal could get through reliably in the presence of other signals that are of different origin (and treated as "noise" with respect to the first signal). Bard Sent via BlackBerry by AT&T From reinoud at castafiore.cde.ua.ac.be Thu Jul 24 11:07:52 2008 From: reinoud at castafiore.cde.ua.ac.be (jim bower) Date: Thu Jul 24 11:13:28 2008 Subject: [Fwd: Re: [Comp-neuro] Noise, redundancy and biophysics] Message-ID: <54471.143.169.8.71.1216890472.squirrel@castafiore.cde.ua.ac.be> ---------------------------- Original Message ---------------------------- Subject: Re: [Comp-neuro] Noise, redundancy and biophysics From: "jim bower" Date: Wed, July 23, 2008 7:49 pm To: "Sacha Nelson" comp-neuro@neuroinf.org -------------------------------------------------------------------------- Not clear that the climbing fiber has any real significance for the output of the cell. For its neutron bomb effect on the dendrite, it produces 1 or 2 spikes as output. My own view is that it is mostly a dendritic phenominon and also has nothing to do with learning. Yes, small cells with no dendrites do something different than large cells with big dendrites, but even the neurons in nucleus lamineris are not well understood. Granule cells have small dendrites, but they stick into a glomerulous which is likely enormously complex. So, generally speaking, generic neuronal processing isn't likely to be found anywhere when you look closely. I am reminded of the first neural network meeting I attended in 1984 at the miramar hotel in Santa Barbara. In fact it was the second meeting of the modern era. And the room was abuzz with talk of the hopfield network and traveling salesman (and spin glasses). I was asked numerous time if there were Hopfield networks in the brain. What I said was no, and therefore it was very likely that artificial neural networks would have to get much more complex and messy to do anything practically interesting. This opinion was definately not well received as the elegance of the Hopfield network, its energy functional, and link to well known theory were seductive properties, and physicists hate complexity. --- 25 years later. - you be the judge. ;-) Jim Sent via BlackBerry by AT&T -----Original Message----- From: Sacha Nelson Date: Wed, 23 Jul 2008 10:54:37 To: Subject: Re: [Comp-neuro] Noise, redundancy and biophysics > Sacha, > > I agree thus it would be nice to agree on a common definition - > which has been very hard to do. > > A couple of questions: > > How can we possibly know what is "relevant" to a particular neuron? > We can decide based on our experimental protocohls, but isn't that > us imposing on "them". I think it is not so much relevant to a particular neuron as relevant to a particular computation. The ultimate arbiter of relevance must always be behavior. Experimentally, it is incredibly hard in most cases to make this causal link: this signal transmitted from this neuron to this other neuron is necessary for the execution of this action or perception, but conceptually I think the acid test is straightforward. I believe that almost any signal that a neurophysiologist could correlate in my brain with some aspect of a stimulus, I could learn to use to discriminate those stimuli and modify my behavior accordingly. But from the point of view of most of my behaviors those signals might be "noise." > > > "the reliability and temporal precision > with which a single presynaptic action potential results in a > postsynaptic action potential can be low". In the cerebellum, the > 150,000 excitatory inputs to each Purkinje cell don't appear to have > any direct influence on the output of the cell - I suspect most > excitatory synaptic inputs in the brain are actually doing exactly > what they appear to be doing, nfluencing the local membrane. I > believe that we neurobiologists may be far too "soma-centric" in > thinking about how neuons and brains compute. yes, but purkinje cells and the apical tufts of pyramidal neurons are somewhat specialized cases. At the other end of the extreme are bushy cells or granule cells. Individual granule cells may not be doing much from the perspective of somatic voltage, but the climbing fiber input certainly is. > > > Last isn't it intersting that the closer one gets to the periphery > either on the sensory or the motor side the more "precise" the > nervous system looks, yet somehow in the middle it looks like it > needs to solve some signal to noise problem. But why isn't it simply > loikely that we can better understand what is going on on either > end, but in the middle we have no idea. I agree it is harder to figure out what is going on in the middle, but it is really about bottlenecks and the relative importance of pure transmission vs. more complicated computation. The retinal ganglion cell axon is actually many synapses away from the photoreceptor, but it is specialized for transmission. Closer to the periphery, exactly what is going on in some amacrine and horizontal cell networks has been incredibly difficult to figure out. The job of the ganglion cell axon (like the climbing fiber) is essentially very different from a recurrent excitatory synapse in the cortex or a granule cell input to a purkinje cell. The experiment of recording from a pre and postsynaptic cell and asking about the reliability, amplitude etc. of transmission can be done anywhere. Ascribing sensory or motor meaning to these signals is of course more difficult, but that is a different business. > > > Although I know bard and others who talk about noise are talking > about it in a precise way -- however, I tend to lump the word > "noise" like a number of other similar words in literature as a bin > in which we put things we don't understand. > > I am reminded of Penzias (sorry about the spelling and wilson > crawling around in their satalite dish with tin foil trying to > remove what turned out to be (predicted) background cosmic radiation. > > (Please don't point out the role of theoretical modeling in > realizing the truth. -- this modeling was done in the context of a > science of simple things (physics) with among other things a common > set of definitions). > > Jim > Sent via BlackBerry by AT&T > From reinoud at castafiore.cde.ua.ac.be Thu Jul 24 11:22:50 2008 From: reinoud at castafiore.cde.ua.ac.be (jim bower) Date: Thu Jul 24 11:31:24 2008 Subject: [Fwd: Re: [Comp-neuro] Noise, redundancy and biophysics] Message-ID: <54511.143.169.8.71.1216891370.squirrel@castafiore.cde.ua.ac.be> ---------------------------- Original Message ---------------------------- Subject: Re: [Comp-neuro] Noise, redundancy and biophysics From: "jim bower" Date: Thu, July 24, 2008 1:00 am To: vibert@u707.jussieu.fr Cc: "Sacha Nelson" comp-neuro@neuroinf.org -------------------------------------------------------------------------- Hi there. Yes I did get your comment previously thank you. As my postings make clear. I am very susipious of the idea of noise in the nervous system. Are your results dependent on "noise" or on some forn of continuous input. And how do you define noise? Jim Sent via BlackBerry by AT&T -----Original Message----- From: Jean-Fran?ois Vibert Date: Wed, 23 Jul 2008 22:24:19 To: Cc: Sacha Nelson; Subject: Re: [Comp-neuro] Noise, redundancy and biophysics Jim, I don't know why, but U receive the mails from the comp_neuro list, but I cannot answer. I wrote an answer 2 days ago, but it was refused by the list (not allowed to write on the list). May be you received it. her is acopy. Other two cents, since we have worked a long time ago, on effect of noise on transfer function of neurons (Segundo, J., Stiber, M., Vibert, J.-F., & Hanneton, S. (1995). Periodically modulated inhibition and its post-synaptic consequences. II. Influence of pre-synaptic slope, depth, range, noise and of post-synaptic natural discharges. Neurosci., 68(3), 693- 719.), and large networks (J. Pham, K. Pakdaman, and J.-F. Vibert, Noise-induced coherent oscillations in randomly connected neural networks. Phys. Rev. E 58, 3610 - 3622 (1998) and written a review on this subject : Segundo, J., Vibert, J.-F., Pakdaman, K., Stiber, M., & Diez Mart?nez, O. (1994). Noise and the neurosciences: A long history, a recent revival and some theory. In K. Pribram (Ed.), Origins: Brain and self organization. Mahwah, NJ: Erlbaum. More recently, we have shown that the respiratory rhythm generation is safe thanks to reticular formation induced noise on ecistatory networks that can be secured by pacemaker neuron if noise is too low (Kosmidis EK, Pierrefiche O, Vibert JF. Respiratory-like rhythmic activity can be produced by an excitatory network of non-pacemaker neuron models. J Neurophysiol. 2004 Aug;92(2):686-99. Regards JF Vibert Le Mer 23 juillet 2008 16:27, jim bower a ?crit : > Sacha, > > I agree thus it would be nice to agree on a common definition - which has > been very hard to do. > > A couple of questions: > > How can we possibly know what is "relevant" to a particular neuron? We > can decide based on our experimental protocohls, but isn't that us > imposing on "them". > > "the reliability and temporal precision > with which a single presynaptic action potential results in a > postsynaptic action potential can be low". In the cerebellum, the 150,000 > excitatory inputs to each Purkinje cell don't appear to have any direct > influence on the output of the cell - I suspect most excitatory synaptic > inputs in the brain are actually doing exactly what they appear to be > doing, nfluencing the local membrane. I believe that we neurobiologists > may be far too "soma-centric" in thinking about how neuons and brains > compute. > > Last isn't it intersting that the closer one gets to the periphery either > on the sensory or the motor side the more "precise" the nervous system > looks, yet somehow in the middle it looks like it needs to solve some > signal to noise problem. But why isn't it simply loikely that we can > better understand what is going on on either end, but in the middle we > have no idea. > > Although I know bard and others who talk about noise are talking about it > in a precise way -- however, I tend to lump the word "noise" like a number > of other similar words in literature as a bin in which we put things we > don't understand. > > I am reminded of Penzias (sorry about the spelling and wilson crawling > around in their satalite dish with tin foil trying to remove what turned > out to be (predicted) background cosmic radiation. > > (Please don't point out the role of theoretical modeling in realizing the > truth. -- this modeling was done in the context of a science of simple > things (physics) with among other things a common set of definitions). > > Jim > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Sacha Nelson > > Date: Wed, 23 Jul 2008 09:40:56 > To: > Cc: > Subject: Re: [Comp-neuro] Noise, redundancy and biophysics > > > It is worth pointing out that "Signal," "noise," and "redundancy" are > highly relative terms. From the perspective of recognizing the word, > adding the missing "a" would be redundnt, while from the perspective > of telling whether or not I can spell it is not. The decoding problem > faced by neurons is really one of separating signals relevant to a > particular computation from signals about other events not relevant to > that computation, but these other signals are not necessarily noise in > the classical (e.g. thermal) sense. > > In nearly every case in which it has been carefully looked at, neurons > turn out to be exquisitely sensitive (e.g relative to behavior). > Central synaptic connections, at least in the adult, turn out to be > far more reliable than once thought. Much of the confusion about noise > and redundancy can be traced to arguments based on systems (e.g. the > neocortex of mammals) where the reliability and temporal precision > with which a single presynaptic action potential results in a > postsynaptic action potential can be low. But this is not a hardware > limitation imposing intrinsic noise, it is a feature of the coding > scheme, common to many central circuits, in which each neuron receives > a very large number (e.g. ~10K) of individual inputs. The ability to > precisely follow a single input is sacrificed in order to be able to > detect correlation (and possibly higher moments) across multiple > inputs. Precisely the opposite end of the spectrum is exemplified in > many circuits closer to the sensory periphery (retinothalamic, > auditory brainstem etc.) or motor output (neuromuscular junction). > Here, hundreds of synaptic boutons comprise an individual connection > and the temporal precision with which in input spike can give rise to > an output spike can be measured in microseconds. Yet, the basic > biophysical properties of the individual synapses and action potential > encoding are essentially the same, or occupy the same range of > properties, in both kinds of circuits. > > For the most part, in large brains at least, the circuit level seems > fairly well insulated from "noise" at the molecular level (i.e. the > stochasticity of channel openings). Of course there are some important > exceptions, and this is less likely to hold in much smaller neurons in > smaller circuits--e.g. worms which only have 300+ neurons and that > lack action potentials entirely. > > -Sacha > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > -- Dr Jean-Fran?ois Vibert; B3E ESIM INSERM UMR-S 707 Facult? de M?decine Piere et Marie Curie sit Saint-Antoine, 27 rue Chaligny. 75571 PARIS Cedex 12 Tel: (+33) 01-44-73-84-31; Fax: (+33) 01-44-73-84-54 e-mail Internet: vibert@u707.jussieu.fr http://www.u707.upmc.fr From zhaoyanchang at hotmail.com Thu Jul 24 12:32:47 2008 From: zhaoyanchang at hotmail.com (Yanchang Zhao) Date: Thu Jul 24 14:09:36 2008 Subject: [Comp-neuro] Final CFP - DDDM 2008, In conjunction with ICDM'08 Message-ID: ******************************************************************** Final Call for Papers - DDDM 2008 The 2nd International Workshop on Domain Driven Data Mining Pisa, Italy, December 15, 2008 In conjunction with IEEE ICDM'08 URL: http://datamining.it.uts.edu.au/dddm08/******************************************************************** The Second International Workshop on Domain Driven Data Mining (DDDM 2008) aims to provide a premier forum for sharing findings, knowledge, insight, experience and lessons in tackling potential challenges in discovering actionable knowledge from complex domain problems, promote the interaction of and fill the gap between data mining research and business expectations, and drive a paradigm shift from traditional data-centered hidden pattern mining to domain-driven actionable knowledge discovery. Submission Instructions-----------------------Paper submissions should be limited to a maximum of 10 pages in the IEEE 2-column format, the same as the camera-ready format (see the IEEE Computer Society Press Proceedings Author Guidelines at http://www.computer.org/portal/pages/cscps/cps/final/icdm06.xml). All papers accepted for the workshop will be included in the ICDM'08 Workshop Proceedings published by the IEEE Computer SocietyPress. Selected papers from the workshop will be invited for consideration of publication in a planned special issue of IEEE Transactions on Knowledge and Data Engineering (subject to approvalfrom TKDE). Important dates--------------- August 1, 2008: Submission deadline September 15, 2008: Notification of paper acceptance to authors October 7, 2008: Deadline for camera-ready copies December 15, 2008: Workshop day Organizing Committee --------------------General Chair Philip S. Yu, University of Illinois at Chicago, USAProgram Chairs Yanchang Zhao, University of Technology, Sydney, Australia Graham Williams, Australian Taxation Office, Australia Carlos Soares, University of Porto, Portugal Contact-------Inquiries can be forwarded to dddm08@it.uts.edu.au. _________________________________________________________________ Windows Live Messenger?treats you to 30 free emoticons -?Bees, cows, tigers and more! http://livelife.ninemsn.com.au/article.aspx?id=567534 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/ce332859/attachment-0001.html From braun at staff.uni-marburg.de Thu Jul 24 14:02:08 2008 From: braun at staff.uni-marburg.de (Hans A. Braun) Date: Thu Jul 24 14:09:40 2008 Subject: [Comp-neuro] Noise etc., comp-neuro mails Message-ID: <005f01c8ed85$198810c0$81c9f889@braun03> Hi Jim, Nathan and everybody, my first mail which was sent on Jim's reply on your, Nathan's announcement has obviously not been distributed via the comp-neuro mailing list. As I have seen, JF Vibert had the same problem. In between, I have been going to the NeuroComp website and registrated again. Maybe, this one works. I am not sure, so I am again have addressed this mail directly to you (few more addresses added). I am pleased to see that such an intense discussion developed. Many interesting points have been made about oscillations and noise. Thanks also for the reference list, Malcolm. Will this be updated? Of course, one should not forget "noise heroes", as for example Frank Moss and Peter Haenggi. And there are many others who have made major contributions. . Especially, I noticed that there is broad agreement that one has to be very careful in comparing biological systems with technical systems. This is the point, very simple, which I wanted to make! Major aspects have perfectly been summarized by Ali. I completely agree, Jim, when you now say that the nervous system uses the technically undesired properties like noise, oscillations etc. as sources of its efficiency. So, it's not the question "whether" but "how". Be careful to ask "why". This often comes close to teleology, wwith implicit reference to a goal motivated, evolutionary engineer, like god. I regret, that I actually don't have much time for more intense discussions. I am under pressure with submitting a paper about noise and oscillations in neural synchronization ;-). Enjoy the hopefully nicely modulated noise from the waves at the beach, Jim. Best regrads to everybody Hans -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/8d258e4c/attachment.html From reinoud at tnb.ua.ac.be Thu Jul 24 14:21:10 2008 From: reinoud at tnb.ua.ac.be (comp-neuro moderator) Date: Thu Jul 24 14:21:34 2008 Subject: [Comp-neuro] rejected msgs Message-ID: <54725.143.169.8.71.1216902070.squirrel@castafiore.cde.ua.ac.be> Dear subscribers, In response to many complaints about rejected messages, please pay attention to the following. The mailing program automatically rejects msgs sent from addresses not in its database. People frequently change address, but keep receiving comp-neuro mail when post is automatically forwarded from their old addresses. Hence receiving comp-neuro mail at a given address is no guarantee that the address is a subscription address, and hence could be used for posting. The only remedy is to repost from the subscription address, or to change the subscription address following instructions at www.neuroinf.org (link to maillists). Sorry for the inconvenience, the comp-neuro moderator. P.S. The moderator will be abroad some days during next two weeks, which may delay broadcasting. From R.Borisyuk at plymouth.ac.uk Thu Jul 24 14:47:32 2008 From: R.Borisyuk at plymouth.ac.uk (Roman Borisyuk) Date: Thu Jul 24 15:21:01 2008 Subject: [Comp-neuro] New paper on development of neural connectivity with particular functionality In-Reply-To: <4870E041.80402@epfl.ch> References: <4870E041.80402@epfl.ch> Message-ID: <6CD19ED93A7A8F4593955A11621242C208EC8B2A7B@ILS133.uopnet.plymouth.ac.uk> New paper "Stochasticity and functionality of neural systems: Mathematical modelling of axon growth in the spinal cord of tadpole" by R. Borisyuk, T. Cook, and A. Roberts has been published in BioSytems, 93, (2008), 101-114. In this paper we study a simple mathematical model of axon growth in the spinal cord of tadpole. The model is relatively fresh and allows fitting to the experimental measurements. Taking into account experimental data on distribution of cell bodies and dendrites along the spinal cord we generate a biologically realistic architecture of the whole tadpole spinal cord. Preliminary study of the electrophysiological properties of the model with Morris-Lecar neurons shows that the model can generate electrical activity corresponding to the swimming pattern of the tadpole. Roman Borisyuk, DSc, PhD Professor of Computational Neuroscience Centre for Theoretical and Computational Neuroscience University of Plymouth A224, Portland Sq Plymouth, PL4 8AA UK Phone: +44 (0) 1752 232619 Fax: +44 (0) 1752 233349 E-mail: RBorisyuk@plymouth.ac.uk -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/5e9fe94f/attachment.html From maass at igi.tugraz.at Thu Jul 24 14:01:33 2008 From: maass at igi.tugraz.at (Wolfgang Maass) Date: Thu Jul 24 15:21:03 2008 Subject: [Comp-neuro] role of noise in learning Message-ID: <48886F1D.9020801@igi.tugraz.at> I would like to add to your discussion that "noise" is obviously needed for reward-based learning in networks of neurons: If such networks have to learn without a supervisor (which tells the neurons during training when they should fire), they have to explore different ways of responding to a stimulus, until the come across responses that are "rewarded" because they provide good system performance. This exploration would appear as "noise" in most analyses. In fact, one might conjecture that networks of neurons are genetically endowed with the capability to go through rather clever exploration patterns (i.e, particular types of "noise"), in order to enable fast convergence of such reinforcement learning schemes. The role of noise in reward-based learning has been analyzed by a number of people, see #183 on http://www.igi.tugraz.at/maass/publications.html for a very recent contribution (and references to earlier work). -Wolfgang -- Prof. Dr. Wolfgang Maass Institut fuer Grundlagen der Informationsverarbeitung Technische Universitaet Graz Inffeldgasse 16b , A-8010 Graz, Austria Tel.: ++43/316/873-5811 Fax ++43/316/873-5805 http://www.igi.tugraz.at/maass/Welcome.html From minai_ali at yahoo.com Thu Jul 24 15:45:49 2008 From: minai_ali at yahoo.com (Ali Minai) Date: Mon Jul 28 14:55:55 2008 Subject: [Comp-neuro] Re: role of noise in learning In-Reply-To: <48884197.4040108@igi.tugraz.at> Message-ID: <234386.67811.qm@web65502.mail.ac4.yahoo.com> Indeed this is an example of the general method by which biological complex systems learn: Explore and reinforce preferentially. This applies to evolution (of course), developmental models of cognitive learning (e.g., ideomotor models), swarm construction (e.g., termite nests), etc. So two questions: 1. Is it even possible for a non-teleological system (i.e., a system that is no driven by a pre-determined target/goal, but does have an implicitly defined fitness/quality function) embedded in a complex environment to (self-)optimize efficiently without the help of noise? 2. Are some types of systems better at this than others? And if so, are there generic principles underlying this? Ali --------------------------------------------------------------------- Ali A. Minai Associate Professor Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: Ali.Minai@uc.edu ????????? minai_ali@yahoo.com WWW: http://www.ece.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Thu, 7/24/08, Wolfgang Maass wrote: From: Wolfgang Maass Subject: role of noise in learning To: minai_ali@yahoo.com, comp-neuro@neuroinf.org Cc: "Etienne B. Roesch" Date: Thursday, July 24, 2008, 4:47 AM I would like to mention that "noise" is obviously also needed for reinforcement learning in networks of neurons: If such networks have to learn without a supervisor (which tells the neurons when they should fire), they have to explore different ways of responding to a stimulus, until the come across responses that are "rewarded" because they provide good network performance. This exploration would appear as "noise" in most analyses. In fact, one might conjecture that networks of neurons are genetically endowed with the capability to carry out particularly useful exploration patterns (i.e, particular types of "noise"), in order to enable fast convergence of such reinforcement learning schemes. This has been demonstrated by a number of people, see #183 on http://www.igi.tugraz.at/maass/publications.html for a very recent contribution (and references to earlier work). w Ali Minai wrote: > Noise does this and much, much more. It can inject variety, break > symmetry, generate novelty, provide energy, facilitate search, carry > signal, and do many other things. Indeed, the only time noise is really > a problem is when one is trying to do achieve a pre-determined goal > (e.g., following a pre-computed trajectory). Since natural systems - > notably the nervous system - rarely (if ever) try to do this, they > thrive on noise. Perhaps we should give the phenomenon a less pejorative > name. "Noise" signals such a linear mindset:-). > > Ali > > > --------------------------------------------------------------------- > Ali A. Minai > Associate Professor > Associate Head for Electrical Engineering > Department of Electrical & Computer Engineering > University of Cincinnati > Cincinnati, OH 45221-0030 > > Phone: (513) 556-4783 > Fax: (513) 556-7326 > Email: aminai@ececs.uc.edu > minai_ali@yahoo.com > > WWW: http://www.ececs.uc.edu/~aminai/ > > ---------------------------------------------------------------------- > > --- On *Tue, 7/22/08, Etienne B. Roesch //* > wrote: > > From: Etienne B. Roesch > Subject: Re: [Comp-neuro] Review announcement > To: comp-neuro@neuroinf.org > Cc: comp-neuro-bounces@neuroinf.org > Date: Tuesday, July 22, 2008, 11:28 AM > > > Yeah, I am loving the discussion! More, more! > > As an early postdoc, I still have in my working memory the classes I > went through in grad school, and I remember this connectionist > lecturer arguing that noise was actually a good thing for > classifier-like systems (and by extension neural nets, and by > extension plausible neural nets -- which are not classifiers stricto > senso I agree) in that it allows an easier discrimination of the > input in a probabilistic context. Given that redundancy of > information/signal plays a big part in how the brain does the job, > wouldn't noise be a clever mechanism to discriminate > close-to-threshold stimuli? What do you think? > > Best regards, > > > Le 22 juil. 08 ? 17:17, jim bower a ?crit : > >> I am actually in a remote part of brazil at the moment, so limited >> to typing on my blackberry. > > Impressive typing skills, I have to admit. ;-) > > >> However, yes I was curious if a discussion could be induced. That >> was originally what this mailing list was set up for, I know, >> because I started it. ;-). However things have become a bit >> complacent so I figured what the heck. >> >> Again limited in my ability to respond but a couple of things. I >> think as computational neurobiologists or scientists in general, >> we need to be aware of the extent which what we can measure >> (oscillations, synchronous spikes, etc) limits the way we think >> about how things work. Many many years ago now when cortical >> oscillations became more generally interesting to people once >> found in visual cortex we suggested based on our realistic >> cortical models that they were an epiphenomina more (loosly) >> reflecting and underlying mechanism for coordinating communication >> and processing between regions than carriers of any information >> themselves. I continue to believe or set my primary assumption >> that until proven otherwise, every spike is significant for >> something and worse yet so is the lack of a spike. >> (Certainly in digital coding 0s are as important as 1s. >> >> Yes "serious scientists" prefer more constrained and defined >> discussions than this. - but we can easily get lost "drinking our >> own whisky". As a famous computational math-bio guy is fond of >> saying. >> >> ;-) >> >> Truth is all these issues really remain wide open. >> >> But and the big but, no evidence that nature is sloppy or >> unsophisticated. >> >> One last point, the assumption that in fact nature is very >> sophisticated and that the structure of the brain deeply reflects >> a complex, sophisticated function pushes in the direction of first >> building models reflecting that structure, even if you are still >> clueless about function. >> >> I am in brazil teaching at the latin american school for >> computational neuroscience, where realistic modeling lives on. ;-) >> >> Best to all >> >> Jim >> >> Sent via BlackBerry by AT&T > > ----- > Etienne Roesch > Department of Computing > Imperial College > London SW7 2AZ > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > ------------------------------------------------------------------------ > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro -- Prof. Dr. Wolfgang Maass Institut fuer Grundlagen der Informationsverarbeitung Technische Universitaet Graz Inffeldgasse 16b , A-8010 Graz, Austria Tel.: ++43/316/873-5811 Fax ++43/316/873-5805 http://www.igi.tugraz.at/maass/Welcome.html -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/b4a0a42c/attachment-0001.html From cpoon at MIT.EDU Thu Jul 24 15:53:03 2008 From: cpoon at MIT.EDU (Chi-Sang Poon) Date: Mon Jul 28 14:57:19 2008 Subject: [Comp-neuro] role of noise in learning In-Reply-To: <48886F1D.9020801@igi.tugraz.at> Message-ID: <027e01c8ed94$9b54d570$d4072a12@PoonCS> Dear Wolfgang and All, Here is another reference for noise-driven reinforcement learning in the context of adaptive control: IEEE Trans Syst Man Cybern B Cybern. 2001;31(2):173-86.Links A Hebbian feedback covariance learning paradigm for self-tuning optimal control. Young DL, Poon CS. We propose a novel adaptive optimal control paradigm inspired by Hebbian covariance synaptic adaptation, a preeminent model of learning and memory as well as other malleable functions in the brain. The adaptation is driven by the spontaneous fluctuations in the system input and output, the covariance of which provides useful information about the changes in the system behavior. The control structure represents a novel form of associative reinforcement learning in which the reinforcement signal is implicitly given by the covariance of the input-output (I/O) signals. Theoretical foundations for the paradigm are derived using Lyapunov theory and are verified by means of computer simulations. The learning algorithm is applicable to a general class of nonlinear adaptive control problems. This on-line direct adaptive control method benefits from a computationally straightforward design, proof of convergence, no need for complete system identification, robustness to noise and uncertainties, and the ability to optimize a general performance criterion in terms of system states and control signals. These attractive properties of Hebbian feedback covariance learning control lend themselves to future investigations into the computational functions of synaptic plasticity in biological neurons. PMID: 18244780 [PubMed - in process] -----Original Message----- From: comp-neuro-bounces@neuroinf.org [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of Wolfgang Maass Sent: Thursday, July 24, 2008 8:02 AM To: comp-neuro@neuroinf.org Cc: nelson@brandeis.edu Subject: [Comp-neuro] role of noise in learning I would like to add to your discussion that "noise" is obviously needed for reward-based learning in networks of neurons: If such networks have to learn without a supervisor (which tells the neurons during training when they should fire), they have to explore different ways of responding to a stimulus, until the come across responses that are "rewarded" because they provide good system performance. This exploration would appear as "noise" in most analyses. In fact, one might conjecture that networks of neurons are genetically endowed with the capability to go through rather clever exploration patterns (i.e, particular types of "noise"), in order to enable fast convergence of such reinforcement learning schemes. The role of noise in reward-based learning has been analyzed by a number of people, see #183 on http://www.igi.tugraz.at/maass/publications.html for a very recent contribution (and references to earlier work). -Wolfgang -- Prof. Dr. Wolfgang Maass Institut fuer Grundlagen der Informationsverarbeitung Technische Universitaet Graz Inffeldgasse 16b , A-8010 Graz, Austria Tel.: ++43/316/873-5811 Fax ++43/316/873-5805 http://www.igi.tugraz.at/maass/Welcome.html _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From psimen at Math.Princeton.EDU Thu Jul 24 16:38:33 2008 From: psimen at Math.Princeton.EDU (Patrick Simen) Date: Mon Jul 28 14:58:30 2008 Subject: [Comp-neuro] role of noise in learning In-Reply-To: <48886F1D.9020801@igi.tugraz.at> References: <48886F1D.9020801@igi.tugraz.at> Message-ID: This is a fascinating discussion. We have a paper in review that is similar in spirit to this interesting abstract, and a poster highlighting its main points at: http://www.math.princeton.edu/~psimen/CCNC2007poster.pdf. It is based on a much coarser-grained model, and is intended to model choice behavior by animals as arising from an integrating or smoothing process applied to intrinsically "noisy" communication processes between neurons. We analyzed a behavioral choice model built from a simple combination of low-pass filters/leaky-integrators/capacitors coupled together by connections that add Gaussian white noise to the signals they propagate. From this model, we prove that a classic "law" of animal behavior from the instrumental conditioning literature emerges (e.g., the "matching law" and its generalized form). There are several earlier models that have made the same point (e.g., Soltani and Wang, J. Neurosci, 2006; Loewenstein and Seung, PNAS, 2006; just to name two). Our model also makes additional analytical predictions about inter-response times that we believe are novel. The point being that not only do neurons perhaps need to explore the space of connection strengths or transfer functions, but that animals similarly need to explore the space of possible responses in order to find rewarding behaviors. I realize that this modeling level may be a bit too abstract for the tastes of this list's readership, but I find it encouraging that very simple learning principles may apply at such widely different temporal and spatial scales. Best wishes, and many thanks for a fresh perspective on this very interesting topic! --Pat. ------------------------------------------------ Patrick Simen, PhD Research Fellow Program in Applied & Computational Mathematics Center for the Study of Brain, Mind & Behavior Princeton University email: psimen@math.princeton.edu http://www.math.princeton.edu/~psimen ------------------------------------------------ On Thu, 24 Jul 2008, Wolfgang Maass wrote: > I would like to add to your discussion that "noise" is obviously > needed for reward-based learning in networks of neurons: > > If such networks have to learn without a supervisor (which tells the neurons > during training when they should fire), they have to explore different ways > of responding to a stimulus, until the come across responses that are > "rewarded" because they provide good system performance. This exploration > would appear as "noise" in most analyses. In fact, one might conjecture that > networks of neurons are genetically endowed with the capability to go through > rather clever exploration patterns (i.e, particular types of "noise"), in > order to enable fast convergence of such reinforcement learning schemes. > > The role of noise in reward-based learning has been analyzed by a number of > people, see #183 on > http://www.igi.tugraz.at/maass/publications.html > for a very recent contribution (and references to earlier work). > > -Wolfgang > > -- > Prof. Dr. Wolfgang Maass > Institut fuer Grundlagen der Informationsverarbeitung > Technische Universitaet Graz > Inffeldgasse 16b , A-8010 Graz, Austria > Tel.: ++43/316/873-5811 > Fax ++43/316/873-5805 > http://www.igi.tugraz.at/maass/Welcome.html > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > From minai_ali at yahoo.com Thu Jul 24 18:00:49 2008 From: minai_ali at yahoo.com (Ali Minai) Date: Mon Jul 28 14:59:56 2008 Subject: [Fwd: Re: [Comp-neuro] Noise, redundancy and biophysics] In-Reply-To: <54471.143.169.8.71.1216890472.squirrel@castafiore.cde.ua.ac.be> Message-ID: <551641.82446.qm@web65502.mail.ac4.yahoo.com> Jim, your point is well-taken, but in fairness, the question to ask is whether that original - perhaps somewhat misguided - infatuation with the Hopfield network hindered or promoted work on more complex models, leading to the more realistic ones we see today. I think that it helped by bringing in a whole community of very smart people - physicists, computer scientists, mathematicians, engineers - some of whom then took the time to learn the biology and eventually made great contributions closer to the level you like. True, they also created a lot of noise, but haven't we all been agreeing that noise is good?:-) I know, of course, that physicists and mathematicians did neuroscience long before Hopfield networks, but not in the numbers they have since. On balance, I think that elegant if not directly applicable ideas such as Hopfield networks, supervised learning, clustering, ICA, etc., have greatly enriched neuroscience by seeding ideas. Cheers Ali --------------------------------------------------------------------- Ali A. Minai Complex Adaptive Systems Lab Associate Professor Department of Electrical & Computer Engineering University of Cincinnati Cincinnati, OH 45221-0030 Phone: (513) 556-4783 Fax: (513) 556-7326 Email: aminai@ece.uc.edu ????????? minai_ali@yahoo.com WWW: http://www.ece.uc.edu/~aminai/ ---------------------------------------------------------------------- --- On Thu, 7/24/08, jim bower wrote: From: jim bower Subject: [Fwd: Re: [Comp-neuro] Noise, redundancy and biophysics] To: comp-neuro@neuroinf.org Date: Thursday, July 24, 2008, 5:07 AM ---------------------------- Original Message ---------------------------- Subject: Re: [Comp-neuro] Noise, redundancy and biophysics From: "jim bower" Date: Wed, July 23, 2008 7:49 pm To: "Sacha Nelson" comp-neuro@neuroinf.org -------------------------------------------------------------------------- Not clear that the climbing fiber has any real significance for the output of the cell. For its neutron bomb effect on the dendrite, it produces 1 or 2 spikes as output. My own view is that it is mostly a dendritic phenominon and also has nothing to do with learning. Yes, small cells with no dendrites do something different than large cells with big dendrites, but even the neurons in nucleus lamineris are not well understood. Granule cells have small dendrites, but they stick into a glomerulous which is likely enormously complex. So, generally speaking, generic neuronal processing isn't likely to be found anywhere when you look closely. I am reminded of the first neural network meeting I attended in 1984 at the miramar hotel in Santa Barbara. In fact it was the second meeting of the modern era. And the room was abuzz with talk of the hopfield network and traveling salesman (and spin glasses). I was asked numerous time if there were Hopfield networks in the brain. What I said was no, and therefore it was very likely that artificial neural networks would have to get much more complex and messy to do anything practically interesting. This opinion was definately not well received as the elegance of the Hopfield network, its energy functional, and link to well known theory were seductive properties, and physicists hate complexity. --- 25 years later. - you be the judge. ;-) Jim Sent via BlackBerry by AT&T -----Original Message----- From: Sacha Nelson Date: Wed, 23 Jul 2008 10:54:37 To: Subject: Re: [Comp-neuro] Noise, redundancy and biophysics > Sacha, > > I agree thus it would be nice to agree on a common definition - > which has been very hard to do. > > A couple of questions: > > How can we possibly know what is "relevant" to a particular neuron? > We can decide based on our experimental protocohls, but isn't that > us imposing on "them". I think it is not so much relevant to a particular neuron as relevant to a particular computation. The ultimate arbiter of relevance must always be behavior. Experimentally, it is incredibly hard in most cases to make this causal link: this signal transmitted from this neuron to this other neuron is necessary for the execution of this action or perception, but conceptually I think the acid test is straightforward. I believe that almost any signal that a neurophysiologist could correlate in my brain with some aspect of a stimulus, I could learn to use to discriminate those stimuli and modify my behavior accordingly. But from the point of view of most of my behaviors those signals might be "noise." > > > "the reliability and temporal precision > with which a single presynaptic action potential results in a > postsynaptic action potential can be low". In the cerebellum, the > 150,000 excitatory inputs to each Purkinje cell don't appear to have > any direct influence on the output of the cell - I suspect most > excitatory synaptic inputs in the brain are actually doing exactly > what they appear to be doing, nfluencing the local membrane. I > believe that we neurobiologists may be far too "soma-centric" in > thinking about how neuons and brains compute. yes, but purkinje cells and the apical tufts of pyramidal neurons are somewhat specialized cases. At the other end of the extreme are bushy cells or granule cells. Individual granule cells may not be doing much from the perspective of somatic voltage, but the climbing fiber input certainly is. > > > Last isn't it intersting that the closer one gets to the periphery > either on the sensory or the motor side the more "precise" the > nervous system looks, yet somehow in the middle it looks like it > needs to solve some signal to noise problem. But why isn't it simply > loikely that we can better understand what is going on on either > end, but in the middle we have no idea. I agree it is harder to figure out what is going on in the middle, but it is really about bottlenecks and the relative importance of pure transmission vs. more complicated computation. The retinal ganglion cell axon is actually many synapses away from the photoreceptor, but it is specialized for transmission. Closer to the periphery, exactly what is going on in some amacrine and horizontal cell networks has been incredibly difficult to figure out. The job of the ganglion cell axon (like the climbing fiber) is essentially very different from a recurrent excitatory synapse in the cortex or a granule cell input to a purkinje cell. The experiment of recording from a pre and postsynaptic cell and asking about the reliability, amplitude etc. of transmission can be done anywhere. Ascribing sensory or motor meaning to these signals is of course more difficult, but that is a different business. > > > Although I know bard and others who talk about noise are talking > about it in a precise way -- however, I tend to lump the word > "noise" like a number of other similar words in literature as a bin > in which we put things we don't understand. > > I am reminded of Penzias (sorry about the spelling and wilson > crawling around in their satalite dish with tin foil trying to > remove what turned out to be (predicted) background cosmic radiation. > > (Please don't point out the role of theoretical modeling in > realizing the truth. -- this modeling was done in the context of a > science of simple things (physics) with among other things a common > set of definitions). > > Jim > Sent via BlackBerry by AT&T > _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080724/6f97ad4d/attachment-0001.html From nips2008publicity at gmail.com Fri Jul 25 06:37:25 2008 From: nips2008publicity at gmail.com (Antonio Torralba) Date: Mon Jul 28 15:02:28 2008 Subject: [Comp-neuro] NIPS Reminder: Workshop and Minisymposia Proposals Due August 1 Message-ID: This is just a reminder that workshop and minisymposia proposals are due by 23:59 PDT on August 1 2008. For more information, please see: http://nips.cc/Conferences/2008/CallForWorkshops Proposals or questions should be emailed as plain text to nips.workshop@gmail.com (please do not use attachments,Word, postscript, html, or pdf files). -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080725/f3f3aeb7/attachment.html From dtam at unt.edu Fri Jul 25 08:01:40 2008 From: dtam at unt.edu (David Tam) Date: Mon Jul 28 15:03:50 2008 Subject: [Comp-neuro] Noise, redundancy and biophysics] In-Reply-To: <54511.143.169.8.71.1216891370.squirrel@castafiore.cde.ua.ac.be> Message-ID: Noise is merely in the eyes of the beholder. What is noise to one, is signal to another. (E.g., Is spam noise?) If you use it, it is signal. If you don't use it (or throw it out), it is noise. That is the classical definition of noise in communication theory. It all depends on the encoder and the decoder. So if the CNS uses it, it is signal, not noise, even if it is stochastic. So some may define noise based on determinism vs. stochasticity rather than the usual definition of noise in engineering that is based on signal content. But determinism and stochasticity is merely in the eyes of the beholder also -- nothing more than the duality of nature, depending on your perspective. If you are god, everything is deterministic. If you are human, everything is stochastic. Just because the system is probabilistic, it doesn't mean it is noisy. Is Brownian motion noisy? It all depends on what you want to get out of those Brownian motions that makes it noisy or not noisy. So noise is really a moot point. To put it more philosophically, noise is what you "don't care" to have. "Don't care" is not necessary a bad philosophical perspective. There are too many examples of "don't care" terms in math logic truth table or computer science that serve real (and convenient) purposes. The stochastic view of the world is a good example of how we only care about the probability but don't care about the determinism. As such, the stochastic view of the universe (or the brain) is not noisy whatsoever, but rather very predictable based on probability. So do neurons (or the brain) use noise in its computation? If the neurons care about those signals, it is not noise. If they don't care, yes, then it is noise. So the question we should be asking is: Are we assigning functions to those poor neurons irrespective of what they are actually doing? Do they really care, literally? :) David Tam, Ph.D. Associate Professor Dept. of Biological Sciences University of North Texas Denton, TX 76203 940-565-3261 From bower at uthscsa.edu Fri Jul 25 23:31:21 2008 From: bower at uthscsa.edu (jim bower) Date: Mon Jul 28 15:05:06 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com><000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> Message-ID: <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry> One last plea for caution. No matter how convenient "noise" may be for abstract models, one can argue I think pretty convincingly that life itself differentiates itself from a-biological chemical evolution through its use of extreme degrees of structure and context specific organization to "beat" thermodynamics. As such anything that smells gaussian, disordered, something you can average over, or something that is completely generic rather than highly specific to the circumstances at hand (behavioral or computational) should raise concern. The long history of biology is that the closer you look, the more structure, not less structure you see. And everytime I have heard it suggested otherwise, closer inspection has suggested that we simply weren't giving biology enough credit. One quick example, a number of years ago, in fact during the first summer of the first computational neuroscience course in woods hole, a well know investigator of aplesia neurons gave a general MBL talk in which he sought to identify every conductance in the neuron of interest. In addition to the big "important" ones he found several small ones, whose presence he chalked up to sloppyness in protein transcription. Given my predispositions, I suggested that he change the temperature of the bath as aplesia live in tide pools. The next summer he gave a talk on the remarkable resiliance of neuronal function under different ambient temperatures confered by the complex conductances present in the cell membrane. While convenient, many of these "noise" mechanisms strike me as being simply not sophisticated enough. And I will point out now for the last time, that the more sophisticated a coding system, the harder it is likely to be to distinguish signal from "noise". Jim Bower Sent via BlackBerry by AT&T -----Original Message----- From: "rod rinkus" Date: Thu, 24 Jul 2008 00:33:59 To: Cc: Subject: RE: [Comp-neuro] Review announcement _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From nurban at cmu.edu Mon Jul 28 17:06:47 2008 From: nurban at cmu.edu (Nathan Urban) Date: Mon Jul 28 17:28:06 2008 Subject: [Comp-neuro] oscillations synchrony and noise. In-Reply-To: <20080728125600.7859D92C6A1@neuroinf.org> References: <20080728125600.7859D92C6A1@neuroinf.org> Message-ID: <488DE087.6040902@cmu.edu> A few points regarding this very interesting recent discussion. 1) I think that there have been several useful points made regarding the definition of "noise". First, noise can be something that is unpredicted given one's knowledge of the conditions of the experiment - such as the variation of a neuron's firing rate when the experimenter presents "the same" stimulus repeatedly. Clearly, lots is going on for the animal besides the stimulus that the experimenter presents. Thus, variation is top be expected. Second, noise can be something that is undesirable given what one is trying to measure at a given time. These two definitions are often quite closely related, although (as pointed out by Jim's example in the case of cosmic background radiation) this can cause serious problems. The third meaning (and the one that we use in the review) is that noise refers to a fluctuating signal (e.g. an output or an input) that has certain statistical properties - like being well described as having every point drawn randomly from a Gaussian distribution. This third usage of the term is agnostic to the cause of these fluctuations. This is the only definition that makes sense in the context of experiments in which we are delivering noisy inputs to a cell or a whole animal. None of these are at al llike thermal noise - which variation in a variable such as temperature that has a particular distribution (e.g. boltzmann) that is due basic physical laws, but where the value of the variable at any given time is not predictable. This is a sort of non-deterministic or essentially stochastic system. In my mind it is this kind of noise that people usually refer to when they say that neurons are in fact not "noisy". My bias is that neurons are not noisy in this way. I think that the vast majority of spikes copuld be predicted if only we knew all the inputs and all the properties of a neuron. However, in fact, I don't think that it matters a lot whether neurons are noisy in this way because in most "noise" due to our uncertainty about the inputs and properties is so dominant. 2) I agree with Jim that approaching the study of the brain with preconceived notions is dangerous. Thus I think that starting from the view that oscillations must be important is to be avoided, but so is starting from the point of view that oscillations are most likely epiphenomena because so many systems oscillate also is to be avoided. My perspective is that many different brain areas show local field potential oscillations - and in many cases these oscillations changes in a way that is correlated with changes in state or behavior of the animal. If we are going to determine whether oscillations are useful in coding, or are analogous to changes in local temperature, we must determine the mechanism of these oscillations so that we can disrupt/augment them. That is, in my mind we must test empirically whether these phenomena functionally interesting or not and the only way to do this is to understand their mechanism. My gut reaction is that in some circumstances the brain cares about the relative timing of spikes across neurons, and that in some cases synchronous spikes are more effective in depolarizing postsynaptic targets than non-synchronous spikes. Thus, I am interested in mechanisms that allow neurons to coordinate their activity at short time scales - such as the noise-induced synchronization that we have described. Nathan -- ____________________________________________ Nathan Urban, Ph.D. Associate Professor Department of Biological Sciences and Center for the Neural Basis of Cognition Carnegie Mellon University 4400 Fifth Ave Pittsburgh PA 15213 ph. 412-268-5122 fax 412-268-8423 http://www.andrew.cmu.edu/user/nurban/Lab_pages/ From ajmandell at charter.net Mon Jul 28 17:20:51 2008 From: ajmandell at charter.net (ajmandell) Date: Mon Jul 28 17:28:11 2008 Subject: [Comp-neuro] Noise stabilization of "unstable fixed point" neuronal firing Message-ID: <002901c8f0c5$855edf70$1b5f2160@shambala> What an interesting discussion! Thought I'd throw in this counterintuitive analogy to the role of noise in conventional stochastic resonance in a neural membrane-like nonlinear dynamical system would be of interest. (1) Real neuronal firing pattern imbedding and poincare section demonstrating stable manifold "gathering" of orbits along the unstable manifold So, P. Francis, J. T. Netoff, T. I.Gluckman, B. J.Schiff, S. J. (Unstable) Periodic orbits: a new language for neuronal dynamics; Biophys. J. 74:2776-85, 1998 (2) Noise induced increases of orbital residence times on unstable manifold of a quasi-membrane equation and map (with neurophysiological speculations) Mandell, AJ and Selz, KA . Brain stem neuronal noise and neocortical "resonance" J. Stat. Physics 70:355-373 Cheers Arnold Arnold J. Mandell, M.D. Cielo Institute, Asheville, NC ajmandell@charter.net Clinical Professor of Psychiatry and Human Behavior, Emory University, Atlanta, GA Adjunct Professor of Mathematical Sciences, FAU, Boca Raton, FL Founding Chairman and Professor Emeritus of Psychiatry (Neuroscience, Pharmacology, Mathematics) UCSD, La Jolla, CA MacArthur Prize Fellow Laureate in Theoretical Neuroscience Humboldt Prize Fellow Laureate in Dynamical Systems -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080728/31d083db/attachment.html From tbosse at few.vu.nl Mon Jul 28 16:32:28 2008 From: tbosse at few.vu.nl (Tibor Bosse) Date: Mon Jul 28 17:28:47 2008 Subject: [Comp-neuro] !Deadline Extension!: 2nd Int Workshop on Human Aspects in Ambient Intelligence Message-ID: [Apologies for multiple copies] ! The submission deadline has been extended to August 8th, see below ! SECOND INTERNATIONAL WORKSHOP ON HUMAN ASPECTS IN AMBIENT INTELLIGENCE: Agent Technology, Human-Oriented Knowledge and Applications (HAI'08) URL: http://www.few.vu.nl/~treur/HAI08wsCfP.htm Sydney, Australia, December 9, 2008 (Financial support for travelling is available, see below) Workshop at the International Conference on Intelligent Agent Technology (IAT'08) Call for Papers Background Recent developments within Ambient Intelligence and Agent Technology provide new possibilities to contribute to personal care. For example, an intelligent ambient agent in our car may monitor us and warn us when we are falling asleep while driving or take measures when we are too drunk to drive. As another example, an elderly person may wear a device with an ambient agent that monitors his or her wellbeing and generates an action when a dangerous situation is noticed. Such Ambient Intelligence applications can be based on the one hand on possibilities to acquire sensor information about humans and their functioning, but on the other hand, more knowledgeable applications crucially depend on the availability of adequate knowledge for analysis of such information about human functioning. If such knowledge about human functioning is computationally available in intelligent software/hardware devices in the environment, such ambient agents can show more human-like understanding and contribute to personal care based on this understanding. In recent years, scientific areas focusing on human functioning such as cognitive science, psychology, neuroscience and biomedical sciences have made substantial progress in providing an increased insight in the various physical and mental aspects of human functioning. Although much work still remains to be done, models have been developed for a variety of such aspects and the way in which humans (try to) manage or regulate them. From a more biomedical angle, examples of such aspects are (management of) heart functioning, diabetes, eating regulation disorders, and HIV-infection. From a more psychological and social angle, examples are emotion regulation, attention regulation, addiction management, trust management, stress management, and criminal behaviour management. If models of human processes and their management are represented in a formal and computational format, and incorporated in the human environment monitoring the physical and mental state of the human, then such ambient agents are able to perform a more in-depth analysis of the human's functioning. An ambience is created that has a human-like understanding of humans, based on computationally formalised knowledge from the human-directed disciplines, and that may more effectively affect the state of humans by undertaking in a knowledgeable manner actions that improve their wellbeing and performance. This may concern elderly people and patients, but also humans in highly demanding circumstances or tasks. For example, the workspaces of naval officers may include systems that, among others, track their eye movements and characteristics of incoming stimuli (e.g., airplanes on a radar screen), and use this information in a computational model that is able to estimate where their attention is focussed at. When it turns out that an officer neglects parts of a radar screen, such a system can either indicate this to the person, or arrange on the background that another person or computer system takes care of this neglected part. Aims This workshop series addresses multidisciplinary aspects of Ambient Intelligence and Agent Systems with human-directed disciplines such as psychology, social science, neuroscience and biomedical sciences. The first workshop in the series (HAI'07) took place at the European Conference on Ambient Intelligence (AmI'07), in Darmstadt, Germany, November 2007. The aim of the workshops is to get researchers together from these human-directed disciplines or working on cross connections of Ambient Intelligence with these disciplines. The focus is on the use of knowledge from these disciplines in Ambient Intelligence applications, in order to take care of and support in a knowledgeable manner humans in their daily living in medical, psychological and social respects. The workshop can play an important role, for example, to get modellers in the psychological, neurological, social or biomedical disciplines interested in Ambient Intelligence as a high-potential application area for their models, and, for example, get inspiration for problem areas to be addressed for further developments in their disciplines. From the other side, the workshop may make researchers in Ambient Intelligence, Agent Systems, and Artificial Intelligence more aware of the possibilities to incorporate more substantial knowledge from the psychological, neurological, social and biomedical disciplines in Ambient Intelligence applications. As part of the interaction, specifications may be generated for experiments to be addressed by the human-directed sciences. Some of the areas of interest * human-aware computing * computational modelling of cognitive, neurological, social and biomedical processes for Ambient Intelligence * modelling emotion and mood and their regulation * collecting and analysing histories of behaviour * computational modelling of mindreading, theory of mind * building profiles; user modelling in Ambient Intelligence * sensoring; e.g., tracking physiological states, gaze, body movements, gestures * sensor information integration methods * analysis of sensor information; e.g., voice and skin analysis with respect to emotional states, gesture analysis, heart rate analysis * environmental modelling * situational awareness * model-based reasoning and analysis techniques for Ambient Intelligence * responsive and adaptive systems; machine learning * cognitive agent models * reflective ambient agent architectures * multi-agent system architectures for Ambient Intelligence applications * human interaction with devices * wearable devices for ambient health and wellness monitoring * brain-computer interfacing * analysis and design of applications to care for humans in need of support for physical and mental health; e.g., elderly or psychiatric care, surveillance, penitentiary care, humans in need of strucural medical or psychological care, support for psychotherapeutical/self-help communities * analysis and design of applications to support humans in demanding circumstances and tasks, such as warfare officers, air traffic controllers, crisis and disaster managers, humans in space missions. * evaluation studies * handling aspects of privacy and security; philosophical and ethical aspects Submission and Proceedings Papers can be submitted in the IEEE 2-column format (see the IEEE Computer Society Press Proceedings Author Guidelines, as for the IAT'08 conference). Expected length is from 3 pages (short papers) to 7 pages (regular papers). Double submission is allowed, but inclusion in the proceedings requires that the paper was and is not published elsewhere. For submissions to the main conference IAT'08, it is possible to indicate explicitly that the paper should be considered for the workshop in case of rejection for the main conference. The workshop proceedings will be published by the IEEE Computer Society Press and will be available at the workshop. More submission details are available at the workshop's Website: http://www.few.vu.nl/~treur/HAI08wsCfP.htm Financial Support for Travelling For those presenters at the workshop for whom excessive travelling costs may cause problems, financial support is available. This support may take the form that for a flight ticket above 400 euro, maximally 75% of the total costs of the ticket can be refunded by the workshop organisation (assuming a ticket of reasonable price for the given distance). After acceptance of a paper, this support can be requested for one author of the paper. Registration For every accepted paper at least one author has to register for the WI / IAT-2008 conference. There is no separate workshop registration fee (i.e., only one conference registration covers everything). Important Dates Submission deadline August 8, 2008 Notification September 3, 2008 Camera ready papers September 30, 2008 Workshop December 9, 2008 Coordination Commitee Juan Carlos Augusto (University of Ulster, School of Computing and Mathematics) Tibor Bosse (Vrije Universiteit Amsterdam, Agent Systems Research Group) Cristiano Castelfranchi (CNR Rome, Institute of Cognitive Sciences and Technologies) Diane Cook (Washington State University, USA) Mark Neerincx (TNO Human Factors; Technical University Delft, Man-Machine Interaction) Fariba Sadri (Imperial College, Department of Computing) Jan Treur (contact person, Vrije Universiteit Amsterdam, Agent Systems Research Group) Programme Committee Juan Carlos Augusto (University of Ulster, School of Computing and Mathematics) Marc B?hlen (State University of New York, USA) Tibor Bosse (Vrije Universiteit Amsterdam, Agent Systems Research Group) Antonio Camurri (University of Genoa, InfoMus Lab) Cristiano Castelfranchi (CNR Rome, Institute of Cognitive Sciences and Technologies) Diane Cook (Washington State University, USA) Hao-Hua Chu (National Taiwan University, Ubicomp Lab, Taiwan) Rino Falcone (CNR Rome, Institute of Cognitive Sciences and Technologies) Dirk Heylen (University of Twente, Human Media Interaction) Anthony Jameson (DFKI, Human-Computer Interaction) Judy Kay (University of Sydney, Computer Human Adaptive Interaction, Australia) Peter Leijdekkers (University of Technology Sydney, Mobile Ubiquitous Services & Technologies Group, Australia) Paul Lukowicz (Austrian University for Health Sciences, Medical Informatics and Technology) Silvia Miksch (Danube University Krems, Department of Information and Knowledge Engineering) Jose del Millan (Swiss Federal Institute of Technology in Lausanne EPFL, Research Institute IDIAP, Martigny, Switzerland) Neelam Naikar (Defence Science and Technology Organisation, Centre for Cognitive Work and Safety Analysis, Australia) Tatsuo Nakajima (Waseda University, Distributed and Ubiquitous Computing Lab, Japan) Mark Neerincx (TNO Human Factors; Technical University Delft, Man-Machine Interaction) Toyoaki Nishida (Kyoto University, Department of Intelligence Science and Technology, Japan) Maja Pantic (University of Twente, Human Media Interaction; Imperial College, Department of Computing, Netherlands/UK) Steffen Pauws (Philips Research Europe, Media Interaction Department, Netherlands) Christian Peter (Fraunhofer Institute for Computer Graphics Rostock, Human-Centered Interaction Technologies, Germany) Tomasz M. Rutkowski (RIKEN Brain Science Institute, Laboratory for Advanced Brain Signal Processing, Japan) Fariba Sadri (Imperial College, Department of Computing) Maarten Sierhuis (NASA Ames Research Center, Human-Centered Computing, USA) Elizabeth Sklar (City University of New York, Brooklyn College, Dept of Computer and Information Science) Ron Sun (Rensselaer Polytechnic Institute, Cognitive Science Department) Bruce H. Thomas (University of South Australia Mawson Lakes, Wearable Computer Lab, Australia) Jan Treur (Vrije Universiteit Amsterdam, Agent Systems Research Group) From bower at uthscsa.edu Tue Jul 29 01:32:45 2008 From: bower at uthscsa.edu (jim bower) Date: Tue Jul 29 14:27:14 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com><000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry><002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> Message-ID: <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry> Ross Great - thanks for support from an unexpected source. ;-). While it is completely contrary to the intuition of most experimentalists who are fond of their single neuron data, it is quite likely that individual neurons in at least the mammalian brain don't matter. Although, as I have mentioned before, one must be aware that neurobiologists pick stimuli to maximize the presumption that they do. On the other hand, in the extreme one could find oneself arguing a la Lashley that there is no structure in cortex and everything is completely distributed. This is also clearly not true. I have suspected for some time that the need for communication (as reflected very loosly in cortical oscillations) is what detemines the size of the population of neurons across which different types of computations are implemented but within which, individuals actually don't matter much. These populations are probably analogous to cortical areas. One other point however, is that abstract modelers tend to think of cortex as one thing and so do many neurobiologists who like to think in terms of things like the cortical microcircuit (or as in the case recently of the blue brain project of a cortical column, which actually doesn't exist - but anyway) in fact it is likely that cerebral cortex is a sequence of things. Getting at what cortex does will require that we understand where this sequence starts and ends. And, no time here and on my poor blackberry to discuss this, but if you are trying to sort these things out in the context of the visual system, as most are, you may be starting at the wrong end. It seems to me likely that it is the olfactory system that invented basic cortical circuitry. And there is a rather interesting story emerging there about just how random cortical activity and connectivity is. Jim bower Jim Sent via BlackBerry by AT&T -----Original Message----- From: "Ross Gayler" Date: Tue, 29 Jul 2008 08:07:20 To: Cc: Subject: Re: [Comp-neuro] Review announcement I had avoided weighing into this discussion on the grounds that my interest lies in abstract connectionist models that are quite remote from the physiological detail that has been the focus here. However, I would like to reinforce a comment that Jim Bower made. > the more sophisticated a coding system, the harder it is likely to be to distinguish signal from "noise". Vector Symbolic Architectures are a family of abstract connectionist models that use very high-dimensional vectors to represent complex data structures that might be required for cognition. (Here's a quick challenge question for you. How might the brain represent "The astronomer believes that the little star orbits the big star"?) In VSA these data structures are represented by the pattern of activity over, say, 10,000 neurons (abstracted as a 10,000 dimensional vector). This is a thoroughly distributed representation in which there is no necessary significance to the activity of any individual neuron - it is only the overall pattern of activity that represents. Furthermore, and this is the point I wanted to make, individual patterns of activity are generally indistinguishable from random vectors. It is only the relationship between representing vectors that carries useful information. An external observer cannot decode the representing vectors without knowing what other vectors are stored in the clean-up memory of the system. I take these representations to be a very clear demonstration of Jim's point that in a sophisticated coding system it is harder to distinguish signal from noise. This paper (http://cogprints.org/3983/) provides some pointers into the VSA literature. The most extensive treatment of a specific instantiation of VSA is Tony Plate's book (http://csli-publications.stanford.edu/site/1575864304.html) Ross Gayler -----Original Message----- From: comp-neuro-bounces@neuroinf.org [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of jim bower Sent: Saturday, 26 July 2008 7:31 AM To: rod rinkus; comp-neuro-bounces@neuroinf.org; minai_ali@yahoo.com Cc: comp-neuro@neuroinf.org Subject: [Possible Spam] Re: [Comp-neuro] Review announcement One last plea for caution. No matter how convenient "noise" may be for abstract models, one can argue I think pretty convincingly that life itself differentiates itself from a-biological chemical evolution through its use of extreme degrees of structure and context specific organization to "beat" thermodynamics. As such anything that smells gaussian, disordered, something you can average over, or something that is completely generic rather than highly specific to the circumstances at hand (behavioral or computational) should raise concern. The long history of biology is that the closer you look, the more structure, not less structure you see. And everytime I have heard it suggested otherwise, closer inspection has suggested that we simply weren't giving biology enough credit. One quick example, a number of years ago, in fact during the first summer of the first computational neuroscience course in woods hole, a well know investigator of aplesia neurons gave a general MBL talk in which he sought to identify every conductance in the neuron of interest. In addition to the big "important" ones he found several small ones, whose presence he chalked up to sloppyness in protein transcription. Given my predispositions, I suggested that he change the temperature of the bath as aplesia live in tide pools. The next summer he gave a talk on the remarkable resiliance of neuronal function under different ambient temperatures confered by the complex conductances present in the cell membrane. While convenient, many of these "noise" mechanisms strike me as being simply not sophisticated enough. And I will point out now for the last time, that the more sophisticated a coding system, the harder it is likely to be to distinguish signal from "noise". Jim Bower Sent via BlackBerry by AT&T -----Original Message----- From: "rod rinkus" Date: Thu, 24 Jul 2008 00:33:59 To: Cc: Subject: RE: [Comp-neuro] Review announcement _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From r.gayler at mbox.com.au Tue Jul 29 00:29:59 2008 From: r.gayler at mbox.com.au (Ross Gayler) Date: Tue Jul 29 14:29:30 2008 Subject: [Comp-neuro] Review announcement Message-ID: <002e01c8f101$78697e60$a500a8c0@Corp.BayAdv> I had avoided weighing into this discussion on the grounds that my interest lies in abstract connectionist models that are quite remote from the physiological detail that has been the focus here. However, I would like to reinforce a comment that Jim Bower made. > the more sophisticated a coding system, the harder it is likely to be to distinguish signal from "noise". Vector Symbolic Architectures are a family of abstract connectionist models that use very high-dimensional vectors to represent complex data structures that might be required for cognition. (Here's a quick challenge question for you. How might the brain represent "The astronomer believes that the little star orbits the big star"?) In VSA these data structures are represented by the pattern of activity over, say, 10,000 neurons (abstracted as a 10,000 dimensional vector). This is a thoroughly distributed representation in which there is no necessary significance to the activity of any individual neuron - it is only the overall pattern of activity that represents. Furthermore, and this is the point I wanted to make, individual patterns of activity are generally indistinguishable from random vectors. It is only the relationship between representing vectors that carries useful information. An external observer cannot decode the representing vectors without knowing what other vectors are stored in the clean-up memory of the system. I take these representations to be a very clear demonstration of Jim's point that in a sophisticated coding system it is harder to distinguish signal from noise. This paper (http://cogprints.org/3983/) provides some pointers into the VSA literature. The most extensive treatment of a specific instantiation of VSA is Tony Plate's book (http://csli-publications.stanford.edu/site/1575864304.html) Ross Gayler -----Original Message----- From: comp-neuro-bounces@neuroinf.org [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of jim bower Sent: Saturday, 26 July 2008 7:31 AM To: rod rinkus; comp-neuro-bounces@neuroinf.org; minai_ali@yahoo.com Cc: comp-neuro@neuroinf.org Subject: [Possible Spam] Re: [Comp-neuro] Review announcement One last plea for caution. No matter how convenient "noise" may be for abstract models, one can argue I think pretty convincingly that life itself differentiates itself from a-biological chemical evolution through its use of extreme degrees of structure and context specific organization to "beat" thermodynamics. As such anything that smells gaussian, disordered, something you can average over, or something that is completely generic rather than highly specific to the circumstances at hand (behavioral or computational) should raise concern. The long history of biology is that the closer you look, the more structure, not less structure you see. And everytime I have heard it suggested otherwise, closer inspection has suggested that we simply weren't giving biology enough credit. One quick example, a number of years ago, in fact during the first summer of the first computational neuroscience course in woods hole, a well know investigator of aplesia neurons gave a general MBL talk in which he sought to identify every conductance in the neuron of interest. In addition to the big "important" ones he found several small ones, whose presence he chalked up to sloppyness in protein transcription. Given my predispositions, I suggested that he change the temperature of the bath as aplesia live in tide pools. The next summer he gave a talk on the remarkable resiliance of neuronal function under different ambient temperatures confered by the complex conductances present in the cell membrane. While convenient, many of these "noise" mechanisms strike me as being simply not sophisticated enough. And I will point out now for the last time, that the more sophisticated a coding system, the harder it is likely to be to distinguish signal from "noise". Jim Bower Sent via BlackBerry by AT&T -----Original Message----- From: "rod rinkus" Date: Thu, 24 Jul 2008 00:33:59 To: Cc: Subject: RE: [Comp-neuro] Review announcement _______________________________________________ Comp-neuro mailing list Comp-neuro@neuroinf.org http://www.neuroinf.org/mailman/listinfo/comp-neuro From bower at uthscsa.edu Tue Jul 29 21:04:48 2008 From: bower at uthscsa.edu (jim bower) Date: Thu Jul 31 12:26:54 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <9EBDC684-18D5-4864-BA90-64D2FBBF2557@cs.ucsd.edu> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com><000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry><002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry><9EBDC684-18D5-4864-BA90-64D2FBBF2557@cs.ucsd.edu> Message-ID: <1434452961-1217358354-cardhu_decombobulator_blackberry.rim.net-139962447-@bxe111.bisx.prod.on.blackberry> ). Hi Jim (from a limo in oakland...!): Would you mind pointing me to data suggesting that cortical columns don't exist? That (among many others you've made, of course!) is an interesting point! Garrison Cottrell --------------------------------- Well, this started as questioning over generalized assumptions and a concern with definitions -- so why not? How about this. Can anyone propose a consistent definition of a cortical column? You might want to start by reading Vernon Mouncastle's original paper (1958 I think), in which there was a very clear definition. (Several years ago he also published a review defending, in effect the original definition). If you read closely, there is actually very little evidence for his original definition. However his assertion that columns represent the " fundamental computational unit "of the cerebral cortex of course does persist as the idea is very attractive as a way to simplify and reduce complexty (and thus see previous). Here are some more specific questions: Do columns have boundaries as Mountcastle clealry believed and believes? If not they are strange columns, if they don't then doesn't that imply that the mapping function is continuous and not discrete? What does that imply about a functional unit? Mountcastles original observation was that neurons sampled by a single electrode inserted perpendicular to the surface of the cortex have the same receptive fields and represented the same subclass of tactle receptors in all layers of somatosensory cortex . This of course is not true. Visual cortex however saved the day. While you can patch together a quasi-columnar pattern in V1, if you try hard enough, it becomes harder and harder the further "in" you go. Personally, I believe that the belief in cortical columns is a consequence of the dominance of computational and experimental studies of the visual system in mammals. (The turtle 'visual' cortex isn't organized this way) as well as our innate desire to find some regularly repeatable and defiable simplification. The assertion that V1-like cortical columns are the funamental computational unit of cerebral cortex has real difficulties if you look at the olfactory system. As I suggested earlier, in my view the likely "inventor" of the cerebral cortical style of computation. Forcing columns into olfactory cortex is extremely difficult. For sure there is a radial organization of cerebral cortex, that is clear from looking at the dendrites of pyramidal cells. - but cortical columns with discrete boundaries, and containing neurons with similar receptive fields (and in .ointcastles original conception representing the same types or peripheral receptors or in other words coding the same information about the stimulous, is an excellent example of what I referred to last week as the tyranny of ideas in neuroscience and computational neuroscience. Which will continue in the absence of a formal system of description and definitions based on the structure of the brain itself. Jim Bower Sent via BlackBerry by AT&T -----Original Message----- From: Garrison Cottrell Date: Tue, 29 Jul 2008 10:21:21 To: Subject: Re: [Comp-neuro] Review announcement Hi Jim (from a limo in oakland...!): Would you mind pointing me to data suggesting that cortical columns don't exist? That (among many others you've made, of course!) is an interesting point! g. On Jul 28, 2008, at 4:32 PM, jim bower wrote: > Ross > > Great - thanks for support from an unexpected source. ;-). > > While it is completely contrary to the intuition of most > experimentalists who are fond of their single neuron data, it is > quite likely that individual neurons in at least the mammalian brain > don't matter. Although, as I have mentioned before, one must be > aware that neurobiologists pick stimuli to maximize the presumption > that they do. > > On the other hand, in the extreme one could find oneself arguing a > la Lashley that there is no structure in cortex and everything is > completely distributed. This is also clearly not true. I have > suspected for some time that the need for communication (as > reflected very loosly in cortical oscillations) is what detemines > the size of the population of neurons across which different types > of computations are implemented but within which, individuals > actually don't matter much. These populations are probably analogous > to cortical areas. > > One other point however, is that abstract modelers tend to think of > cortex as one thing and so do many neurobiologists who like to think > in terms of things like the cortical microcircuit (or as in the case > recently of the blue brain project of a cortical column, which > actually doesn't exist - but anyway) in fact it is likely that > cerebral cortex is a sequence of things. Getting at what cortex does > will require that we understand where this sequence starts and ends. > And, no time here and on my poor blackberry to discuss this, but if > you are trying to sort these things out in the context of the visual > system, as most are, you may be starting at the wrong end. It seems > to me likely that it is the olfactory system that invented basic > cortical circuitry. And there is a rather interesting story emerging > there about just how random cortical activity and connectivity is. > > Jim bower > > Jim > Sent via BlackBerry by AT&T > > -----Original Message----- > From: "Ross Gayler" > > Date: Tue, 29 Jul 2008 08:07:20 > To: > Cc: > Subject: Re: [Comp-neuro] Review announcement > > > I had avoided weighing into this discussion on the grounds that my > interest > lies in abstract connectionist models that are quite remote from the > physiological detail that has been the focus here. However, I would > like to > reinforce a comment that Jim Bower made. > >> the more sophisticated a coding system, the harder it is likely to >> be to > distinguish signal from "noise". > > Vector Symbolic Architectures are a family of abstract connectionist > models > that use very high-dimensional vectors to represent complex data > structures > that might be required for cognition. (Here's a quick challenge > question for > you. How might the brain represent "The astronomer believes that the > little > star orbits the big star"?) In VSA these data structures are > represented by > the pattern of activity over, say, 10,000 neurons (abstracted as a > 10,000 > dimensional vector). This is a thoroughly distributed > representation in > which there is no necessary significance to the activity of any > individual > neuron - it is only the overall pattern of activity that represents. > > Furthermore, and this is the point I wanted to make, individual > patterns of > activity are generally indistinguishable from random vectors. It is > only > the relationship between representing vectors that carries useful > information. An external observer cannot decode the representing > vectors > without knowing what other vectors are stored in the clean-up memory > of the > system. I take these representations to be a very clear > demonstration of > Jim's point that in a sophisticated coding system it is harder to > distinguish signal from noise. > > This paper (http://cogprints.org/3983/) provides some pointers into > the VSA > literature. The most extensive treatment of a specific > instantiation of VSA > is Tony Plate's book > (http://csli-publications.stanford.edu/site/1575864304.html) > > > Ross Gayler > > > > > -----Original Message----- > From: comp-neuro-bounces@neuroinf.org > [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of jim bower > Sent: Saturday, 26 July 2008 7:31 AM > To: rod rinkus; comp-neuro-bounces@neuroinf.org; minai_ali@yahoo.com > Cc: comp-neuro@neuroinf.org > Subject: [Possible Spam] Re: [Comp-neuro] Review announcement > > One last plea for caution. No matter how convenient "noise" may be for > abstract models, one can argue I think pretty convincingly that life > itself > differentiates itself from a-biological chemical evolution through > its use > of extreme degrees of structure and context specific organization to > "beat" > thermodynamics. > > As such anything that smells gaussian, disordered, something you can > average > over, or something that is completely generic rather than highly > specific to > the circumstances at hand (behavioral or computational) should raise > concern. > > The long history of biology is that the closer you look, the more > structure, > not less structure you see. And everytime I have heard it suggested > otherwise, closer inspection has suggested that we simply weren't > giving > biology enough credit. > > One quick example, a number of years ago, in fact during the first > summer of > the first computational neuroscience course in woods hole, a well know > investigator of aplesia neurons gave a general MBL talk in which he > sought > to identify every conductance in the neuron of interest. In addition > to the > big "important" ones he found several small ones, whose presence he > chalked > up to sloppyness in protein transcription. > > Given my predispositions, I suggested that he change the temperature > of the > bath as aplesia live in tide pools. The next summer he gave a talk > on the > remarkable resiliance of neuronal function under different ambient > temperatures confered by the complex conductances present in the cell > membrane. > > While convenient, many of these "noise" mechanisms strike me as > being simply > not sophisticated enough. And I will point out now for the last > time, that > the more sophisticated a coding system, the harder it is likely to > be to > distinguish signal from "noise". > > Jim Bower > Sent via BlackBerry by AT&T > > -----Original Message----- > From: "rod rinkus" > > Date: Thu, 24 Jul 2008 00:33:59 > To: > Cc: > Subject: RE: [Comp-neuro] Review announcement > > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro Gary Cottrell 858-534-6640 FAX: 858-534-7029 Computer Science and Engineering 0404 IF USING FED EX INCLUDE THE FOLLOWING LINE: CSE Building, Room 4130 University of California San Diego - 9500 Gilman Drive # 0404 La Jolla, Ca. 92093-0404 "Only connect!" -E.M. Forster "I am awaiting the day when people remember the fact that discovery does not work by deciding what you want and then discovering it." -David Mermin Email: gary@ucsd.edu Home page: http://www-cse.ucsd.edu/~gary/ From r.gayler at gmail.com Wed Jul 30 06:35:33 2008 From: r.gayler at gmail.com (Ross Gayler) Date: Thu Jul 31 12:26:57 2008 Subject: [Comp-neuro] cognitive representations & abstract modelling In-Reply-To: <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com><000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry><002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry> Message-ID: <006501c8f1fd$b6dd3770$a500a8c0@Corp.BayAdv> Jim, (I've changed the subject line to reflect the direction of this post.) > thanks for support from an unexpected source I'm always happy to help someone out of the black pit of grubby physiological detail and into the pure, clear light of abstract models. ;-) But, seriously though ... > if you are trying to sort these things out in the context of the visual system, as most are, you may be starting at the wrong end. It seems to me likely that it is the olfactory system that invented basic cortical circuitry. Lucky for me my models are so abstracted that I don't make any distinction between sensory modalities :-) My focus is on specific representational and computational issues rather than physiological implementation. All I care about in that direction is that implementation of the models as real systems should not be blatantly infeasible. One of the interesting aspects of the cognitive problem domain is the essentially open-ended nature of the conceptual space. New concepts are being created and used all the time. Contrast this with a low level sensory space, say - the retina, where the representation space is fixed by the set of photoreceptors. By analogy, running a visual system like a cognitive sytem would require continually adding new photoreceptors in new locations with new spectral sensitivities. The open-endedness of conceptual space strongly suggests (at least, to me) that the cognitive level of representation is quasi-independent of the physiological implementation. That is, the physiological circuitry implements a virtual machine on which the cognitive processes are executed while remaining relatively independent of what is happening at the physiological level. To give a more concrete example: in Vector Symbolic Architectures you could implement a virtual "grandmother cell" whose implementation is distributed over a large number of actual neurons, none of which is dedicated to only that task. You could also dynamically create virtual cells for "proton spin", "online social networking" and "incestuous love triangle as punishment for peeving the gods" simultaneously implemented over the same actual neurons. (There is a little more about this idea at http://cogprints.org/500/) However, this distributed virtual machine view does not necessarily commit me to "arguing a la Lashley that there is no structure in cortex and everything is completely distributed". I am very partial to Lawrence Barsalou's "Perceptual Symbol Systems" hypothesis (http://www.psychology.emory.edu/cognition/barsalou/onlinepapers.html). He argues that cognition consists of modal perceptual systems being run as simulators. This would be consistent with different cortical areas being modality specialists (with the possibility of modality-specific architectures) provided that there is some communication mechanism to allow integration of the cognitive simulations of the different cortical areas. VSA provides some room for optimism here. It deals only with vectors of activity and is blind to the architecture that generated the activities - so effective communication between cortical areas with different architectures would not be out of the question. However, there is the problem of how one cortical area learns what the input from another area means. Without wishing to appear glib, this is the same problem that a cortical area faces with respect to its own sensory input. Every input is just a vector of activities, so if the architecture can solve the "what does it mean" problem for one input it should be able to do the same for all inputs. In a similar vein, comp-neuro readers may be interested in Rich Sutton's "Predictive State Representations" (http://www.cs.ualberta.ca/~sutton/Talks/Talks.html#Predictive_Representatio ns_of_State_and). PSRs represent the state of an agent's knowledge about its current situation as a set of predicted future sensorimotor interactions with the world. This is a radical claim, as it asserts that *all* knowledge is repesented in sensorimotor terms (an idea that has been around for some time, e.g. see http://cogweb.ucla.edu/Abstracts/Johnson_87.html). PSRs ought to be appealing to neurophysiologists because they are very concretely tied to sensorimotor events. At the same time, they cry out for a virtualising implementation (like VSA) because they are predictions of sequences of events that have not yet happened and because the agent must be able to represent many such sequences simultaneously on the same physiological implementation. So, returning to your initial statement that "it is quite likely that individual neurons in at least the mammalian brain don't matter" - I agree on the grounds that it appears to be a requirement flowing from having a cognitive system. Ross Gayler -----Original Message----- From: jim bower [mailto:bower@uthscsa.edu] Sent: Tuesday, 29 July 2008 9:33 AM To: r.gayler@gmail.com Cc: comp-neuro@neuroinf.org Subject: Re: [Comp-neuro] Review announcement Ross Great - thanks for support from an unexpected source. ;-). While it is completely contrary to the intuition of most experimentalists who are fond of their single neuron data, it is quite likely that individual neurons in at least the mammalian brain don't matter. Although, as I have mentioned before, one must be aware that neurobiologists pick stimuli to maximize the presumption that they do. On the other hand, in the extreme one could find oneself arguing a la Lashley that there is no structure in cortex and everything is completely distributed. This is also clearly not true. I have suspected for some time that the need for communication (as reflected very loosly in cortical oscillations) is what detemines the size of the population of neurons across which different types of computations are implemented but within which, individuals actually don't matter much. These populations are probably analogous to cortical areas. One other point however, is that abstract modelers tend to think of cortex as one thing and so do many neurobiologists who like to think in terms of things like the cortical microcircuit (or as in the case recently of the blue brain project of a cortical column, which actually doesn't exist - but anyway) in fact it is likely that cerebral cortex is a sequence of things. Getting at what cortex does will require that we understand where this sequence starts and ends. And, no time here and on my poor blackberry to discuss this, but if you are trying to sort these things out in the context of the visual system, as most are, you may be starting at the wrong end. It seems to me likely that it is the olfactory system that invented basic cortical circuitry. And there is a rather interesting story emerging there about just how random cortical activity and connectivity is. Jim bower From rinkus at comcast.net Wed Jul 30 17:48:46 2008 From: rinkus at comcast.net (rinkus@comcast.net) Date: Thu Jul 31 12:27:00 2008 Subject: [Comp-neuro] two underlying questions about noise Message-ID: <073020081548.12058.48908D5E0000713E00002F1A22069984999C9A0502079D@comcast.net> Dr. Bower and all, Thanks for this wonderful discussion with such a grand charter on noise and synchrony. I see two different questions re noise underlying the discussion. 1. Is the pattern of firing observed for any particular principal cell (PC) noisy? 2. Is noise (randomness) used by the system in the process of storing information? Regarding the first, given that the typical PC has about 10,000 afferent synapses and that the experimenter knows very little about the simultaneous state of those 10,000 inputs, this is a hard question to answer with certainty. I agree that we'd see a lot less, and possibly very little, randomness in the PC's firing pattern if we had complete knowledge of its 10,000 inputs. Regarding the second question, in my earlier post, I described a simple mechanism whereby noise is used by the system to ensure that a certain property--namely, that similar inputs map to similar codes (SISC)--exists over the set of codes stored in a given patch of cortex. By 'codes', I mean specifically, 'sparse distributed codes'. My working hypothesis is that the neuromodulatory systems of NE and/or Ach might instantiate this noise mechanism (cf. work by Isakova, Kimura, Tateno, Hasselmo, Vankov, Clayton, Dayan, Bouret, Sara, and many others). When the input to the patch is detected to be novel, one/both of NE/Ach are flooded into the patch, swamping out the effects of specific synaptic inputs, resulting in choosing a small set of PC's in the patch essentially at random. However, if we could immediately give the same input again to the patch, no novelty (perfect familiarity) would be detected, thus no Ach/NE would be flooded to the patch and the specific synaptic inputs would dominate, which would with very high probability (with little randomness) cause the same set of PC's (code) to become active (recognition). Thus, in my model, noise (manifest as momentary Ach/NE levels) is most definitely used during construction of memories. However, after construction, if we could test, with complete control over the PC's inputs, the firing pattern would show very little noise. -Rod Rinkus From darioringach at mac.com Thu Jul 31 15:56:09 2008 From: darioringach at mac.com (Dario Ringach) Date: Fri Aug 1 13:09:44 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <1434452961-1217358354-cardhu_decombobulator_blackberry.rim.net-139962447-@bxe111.bisx.prod.on.blackberry> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com> <000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry> <002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry> <9EBDC684-18D5-4864-BA90-64D2FBBF2557@cs.ucsd.edu> <1434452961-1217358354-cardhu_decombobulator_blackberry.rim.net-139962447-@bxe111.bisx.prod.on.blackberry> Message-ID: A particular interesting argument about the function (actually, the lack thereof) of cortical column is: Horton JC, Adams DL. The cortical column: a structure without a function. Philos Trans R Soc Lond B Biol Sci. 2005 Apr 29;360(1456):837-62. -- Dario Dario Ringach, PhD Professor of Neurobiology and Psychology Jules Stein Eye Institute David Geffen School of Medicine University of California, Los Angeles dario@ucla.edu | http://web.mac.com/darioringach On Jul 29, 2008, at 12:04 PM, jim bower wrote: > ). Hi Jim (from a limo in oakland...!): > > Would you mind pointing me to data suggesting that cortical columns > don't exist? That (among many others you've made, of course!) is an > interesting point! > > Garrison Cottrell > > --------------------------------- > Well, this started as questioning over generalized assumptions and a > concern with definitions -- so why not? > > How about this. Can anyone propose a consistent definition of a > cortical column? > > You might want to start by reading Vernon Mouncastle's original > paper (1958 I think), in which there was a very clear definition. > (Several years ago he also published a review defending, in effect > the original definition). > > If you read closely, there is actually very little evidence for his > original definition. However his assertion that columns represent > the " fundamental computational unit "of the cerebral cortex of > course does persist as the idea is very attractive as a way to > simplify and reduce complexty (and thus see previous). > > Here are some more specific questions: > > Do columns have boundaries as Mountcastle clealry believed and > believes? If not they are strange columns, if they don't then > doesn't that imply that the mapping function is continuous and not > discrete? What does that imply about a functional unit? > > Mountcastles original observation was that neurons sampled by a > single electrode inserted perpendicular to the surface of the cortex > have the same receptive fields and represented the same subclass of > tactle receptors in all layers of somatosensory cortex . This of > course is not true. > > Visual cortex however saved the day. > > While you can patch together a quasi-columnar pattern in V1, if you > try hard enough, it becomes harder and harder the further "in" you go. > > Personally, I believe that the belief in cortical columns is a > consequence of the dominance of computational and experimental > studies of the visual system in mammals. (The turtle 'visual' cortex > isn't organized this way) as well as our innate desire to find some > regularly repeatable and defiable simplification. > > The assertion that V1-like cortical columns are the funamental > computational unit of cerebral cortex has real difficulties if you > look at the olfactory system. As I suggested earlier, in my view the > likely "inventor" of the cerebral cortical style of computation. > Forcing columns into olfactory cortex is extremely difficult. For > sure there is a radial organization of cerebral cortex, that is > clear from looking at the dendrites of pyramidal cells. - but > cortical columns with discrete boundaries, and containing neurons > with similar receptive fields (and in .ointcastles original > conception representing the same types or peripheral receptors or in > other words coding the same information about the stimulous, is an > excellent example of what I referred to last week as the tyranny of > ideas in neuroscience and computational neuroscience. Which will > continue in the absence of a formal system of description and > definitions based on the structure of the brain itself. > > Jim Bower > > > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Garrison Cottrell > > Date: Tue, 29 Jul 2008 10:21:21 > To: > Subject: Re: [Comp-neuro] Review announcement > > > Hi Jim (from a limo in oakland...!): > > Would you mind pointing me to data suggesting that cortical columns > don't exist? That (among many others you've made, of course!) is an > interesting point! > > g. > > On Jul 28, 2008, at 4:32 PM, jim bower wrote: > >> Ross >> >> Great - thanks for support from an unexpected source. ;-). >> >> While it is completely contrary to the intuition of most >> experimentalists who are fond of their single neuron data, it is >> quite likely that individual neurons in at least the mammalian brain >> don't matter. Although, as I have mentioned before, one must be >> aware that neurobiologists pick stimuli to maximize the presumption >> that they do. >> >> On the other hand, in the extreme one could find oneself arguing a >> la Lashley that there is no structure in cortex and everything is >> completely distributed. This is also clearly not true. I have >> suspected for some time that the need for communication (as >> reflected very loosly in cortical oscillations) is what detemines >> the size of the population of neurons across which different types >> of computations are implemented but within which, individuals >> actually don't matter much. These populations are probably analogous >> to cortical areas. >> >> One other point however, is that abstract modelers tend to think of >> cortex as one thing and so do many neurobiologists who like to think >> in terms of things like the cortical microcircuit (or as in the case >> recently of the blue brain project of a cortical column, which >> actually doesn't exist - but anyway) in fact it is likely that >> cerebral cortex is a sequence of things. Getting at what cortex does >> will require that we understand where this sequence starts and ends. >> And, no time here and on my poor blackberry to discuss this, but if >> you are trying to sort these things out in the context of the visual >> system, as most are, you may be starting at the wrong end. It seems >> to me likely that it is the olfactory system that invented basic >> cortical circuitry. And there is a rather interesting story emerging >> there about just how random cortical activity and connectivity is. >> >> Jim bower >> >> Jim >> Sent via BlackBerry by AT&T >> >> -----Original Message----- >> From: "Ross Gayler" >> >> Date: Tue, 29 Jul 2008 08:07:20 >> To: >> Cc: >> Subject: Re: [Comp-neuro] Review announcement >> >> >> I had avoided weighing into this discussion on the grounds that my >> interest >> lies in abstract connectionist models that are quite remote from the >> physiological detail that has been the focus here. However, I would >> like to >> reinforce a comment that Jim Bower made. >> >>> the more sophisticated a coding system, the harder it is likely to >>> be to >> distinguish signal from "noise". >> >> Vector Symbolic Architectures are a family of abstract connectionist >> models >> that use very high-dimensional vectors to represent complex data >> structures >> that might be required for cognition. (Here's a quick challenge >> question for >> you. How might the brain represent "The astronomer believes that the >> little >> star orbits the big star"?) In VSA these data structures are >> represented by >> the pattern of activity over, say, 10,000 neurons (abstracted as a >> 10,000 >> dimensional vector). This is a thoroughly distributed >> representation in >> which there is no necessary significance to the activity of any >> individual >> neuron - it is only the overall pattern of activity that represents. >> >> Furthermore, and this is the point I wanted to make, individual >> patterns of >> activity are generally indistinguishable from random vectors. It is >> only >> the relationship between representing vectors that carries useful >> information. An external observer cannot decode the representing >> vectors >> without knowing what other vectors are stored in the clean-up memory >> of the >> system. I take these representations to be a very clear >> demonstration of >> Jim's point that in a sophisticated coding system it is harder to >> distinguish signal from noise. >> >> This paper (http://cogprints.org/3983/) provides some pointers into >> the VSA >> literature. The most extensive treatment of a specific >> instantiation of VSA >> is Tony Plate's book >> (http://csli-publications.stanford.edu/site/1575864304.html) >> >> >> Ross Gayler >> >> >> >> >> -----Original Message----- >> From: comp-neuro-bounces@neuroinf.org >> [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of jim bower >> Sent: Saturday, 26 July 2008 7:31 AM >> To: rod rinkus; comp-neuro-bounces@neuroinf.org; minai_ali@yahoo.com >> Cc: comp-neuro@neuroinf.org >> Subject: [Possible Spam] Re: [Comp-neuro] Review announcement >> >> One last plea for caution. No matter how convenient "noise" may be >> for >> abstract models, one can argue I think pretty convincingly that life >> itself >> differentiates itself from a-biological chemical evolution through >> its use >> of extreme degrees of structure and context specific organization to >> "beat" >> thermodynamics. >> >> As such anything that smells gaussian, disordered, something you can >> average >> over, or something that is completely generic rather than highly >> specific to >> the circumstances at hand (behavioral or computational) should raise >> concern. >> >> The long history of biology is that the closer you look, the more >> structure, >> not less structure you see. And everytime I have heard it suggested >> otherwise, closer inspection has suggested that we simply weren't >> giving >> biology enough credit. >> >> One quick example, a number of years ago, in fact during the first >> summer of >> the first computational neuroscience course in woods hole, a well >> know >> investigator of aplesia neurons gave a general MBL talk in which he >> sought >> to identify every conductance in the neuron of interest. In addition >> to the >> big "important" ones he found several small ones, whose presence he >> chalked >> up to sloppyness in protein transcription. >> >> Given my predispositions, I suggested that he change the temperature >> of the >> bath as aplesia live in tide pools. The next summer he gave a talk >> on the >> remarkable resiliance of neuronal function under different ambient >> temperatures confered by the complex conductances present in the cell >> membrane. >> >> While convenient, many of these "noise" mechanisms strike me as >> being simply >> not sophisticated enough. And I will point out now for the last >> time, that >> the more sophisticated a coding system, the harder it is likely to >> be to >> distinguish signal from "noise". >> >> Jim Bower >> Sent via BlackBerry by AT&T >> >> -----Original Message----- >> From: "rod rinkus" >> >> Date: Thu, 24 Jul 2008 00:33:59 >> To: >> Cc: >> Subject: RE: [Comp-neuro] Review announcement >> >> >> _______________________________________________ >> Comp-neuro mailing list >> Comp-neuro@neuroinf.org >> http://www.neuroinf.org/mailman/listinfo/comp-neuro >> >> >> _______________________________________________ >> Comp-neuro mailing list >> Comp-neuro@neuroinf.org >> http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > Gary Cottrell 858-534-6640 FAX: 858-534-7029 > Computer Science and Engineering 0404 > IF USING FED EX INCLUDE THE FOLLOWING LINE: > CSE Building, Room 4130 > University of California San > Diego - > 9500 Gilman Drive # 0404 > La Jolla, Ca. 92093-0404 > > "Only connect!" -E.M. Forster > > "I am awaiting the day when people remember the fact that discovery > does not work by deciding what you want and then discovering it." > -David Mermin > > > Email: gary@ucsd.edu > Home page: http://www-cse.ucsd.edu/~gary/ > > > > _______________________________________________ > Comp-neuro mailing list > Comp-neuro@neuroinf.org > http://www.neuroinf.org/mailman/listinfo/comp-neuro -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080731/8c810342/attachment-0001.html From fabio_mss at hotmail.com Thu Jul 31 20:29:21 2008 From: fabio_mss at hotmail.com (=?Windows-1252?Q?Fabio_Marques_Sim=F5es_de_Souza?=) Date: Fri Aug 1 13:09:50 2008 Subject: [Comp-neuro] Review announcement In-Reply-To: <1434452961-1217358354-cardhu_decombobulator_blackberry.rim.net-139962447-@bxe111.bisx.prod.on.blackberry> References: <989871.36961.qm@web65515.mail.ac4.yahoo.com><000701c8ed46$7e9d97d0$0302a8c0@DGF7K891> <243070410-1217021545-cardhu_decombobulator_blackberry.rim.net-58570633-@bxe111.bisx.prod.on.blackberry><002d01c8f0fe$55aade80$a500a8c0@Corp.BayAdv> <343959222-1217288029-cardhu_decombobulator_blackberry.rim.net-1542841324-@bxe111.bisx.prod.on.blackberry><9EBDC684-18D5-4864-BA90-64D2FBBF2557@cs.ucsd.edu> <1434452961-1217358354-cardhu_decombobulator_blackberry.rim.net-139962447-@bxe111.bisx.prod.on.blackberry> Message-ID: Hi Jim, I am glad you are enjoying your stay there in Brazil! I am following this nice scientific discussion. I have some points to add: Cortical columns do exist, but they have no clear function -if any-. The following review is discussing this issue: Horton J C and Adams D L The cortical column: a structure without a function. Phil. Trans. R. Soc. (2005) 360: 837-862. "This year, the field of neuroscience celebrates the 50th anniversary of Mountcastle?s discovery of the cortical column. In this review,we summarize half a century of research and come to the disappointing realization that the column may have no function...". Regarding the olfactory system and noisy single unit coding in general: The olfactory bulb is the very first brain structure that get inputs from the receptor layer, and, as you know, it is very well organized in discrete glomeruli that receive input of receptor cells expressing a single receptor protein. Do glomeruli have a function? The bulbs are telencephalic structures, and I am wondering if the cortical columns may be homologous to glomeruli of the olfactory bulbs. It is just hard to believe that evolution would preserve structures like cortical columns and glomeruli if they had no function at all. Regarding noisy single unit coding, there is no doubt there is signal there, otherwise researchers never would be able to extract signal from a hundred of single units spread in the motor cortex in a way that allows a monkey to move a robotic arm in real time using its own brain in the same way it would move his own arm. Is it valid only for motor cortex that is close to the "output" of the system? How to explain it? Does each area of the brain has its own way to compute information or there are few basic principles that apply for the whole brain? Would be the cortical oscillations reflecting a type of "clock" that drives brain activity? In other words, can LFPs reflect the speed the brain is computing information, but not the code itself? Where is the code and how to crack it? Tudo de bom! Fabio > To: gary@cs.ucsd.edu; comp-neuro@neuroinf.org > Subject: Re: [Comp-neuro] Review announcement > From: bower@uthscsa.edu > Date: Tue, 29 Jul 2008 19:04:48 +0000 > CC: > > ). Hi Jim (from a limo in oakland...!): > > Would you mind pointing me to data suggesting that cortical columns > don't exist? That (among many others you've made, of course!) is an > interesting point! > > Garrison Cottrell > > --------------------------------- > Well, this started as questioning over generalized assumptions and a concern with definitions -- so why not? > > How about this. Can anyone propose a consistent definition of a cortical column? > > You might want to start by reading Vernon Mouncastle's original paper (1958 I think), in which there was a very clear definition. (Several years ago he also published a review defending, in effect the original definition). > > If you read closely, there is actually very little evidence for his original definition. However his assertion that columns represent the " fundamental computational unit "of the cerebral cortex of course does persist as the idea is very attractive as a way to simplify and reduce complexty (and thus see previous). > > Here are some more specific questions: > > Do columns have boundaries as Mountcastle clealry believed and believes? If not they are strange columns, if they don't then doesn't that imply that the mapping function is continuous and not discrete? What does that imply about a functional unit? > > Mountcastles original observation was that neurons sampled by a single electrode inserted perpendicular to the surface of the cortex have the same receptive fields and represented the same subclass of tactle receptors in all layers of somatosensory cortex . This of course is not true. > > Visual cortex however saved the day. > > While you can patch together a quasi-columnar pattern in V1, if you try hard enough, it becomes harder and harder the further "in" you go. > > Personally, I believe that the belief in cortical columns is a consequence of the dominance of computational and experimental studies of the visual system in mammals. (The turtle 'visual' cortex isn't organized this way) as well as our innate desire to find some regularly repeatable and defiable simplification. > > The assertion that V1-like cortical columns are the funamental computational unit of cerebral cortex has real difficulties if you look at the olfactory system. As I suggested earlier, in my view the likely "inventor" of the cerebral cortical style of computation. Forcing columns into olfactory cortex is extremely difficult. For sure there is a radial organization of cerebral cortex, that is clear from looking at the dendrites of pyramidal cells. - but cortical columns with discrete boundaries, and containing neurons with similar receptive fields (and in .ointcastles original conception representing the same types or peripheral receptors or in other words coding the same information about the stimulous, is an excellent example of what I referred to last week as the tyranny of ideas in neuroscience and computational neuroscience. Which will continue in the absence of a formal system of description and definitions based on the structure of the brain itself. > > Jim Bower > > > Sent via BlackBerry by AT&T > > -----Original Message----- > From: Garrison Cottrell > > Date: Tue, 29 Jul 2008 10:21:21 > To: > Subject: Re: [Comp-neuro] Review announcement > > > Hi Jim (from a limo in oakland...!): > > Would you mind pointing me to data suggesting that cortical columns > don't exist? That (among many others you've made, of course!) is an > interesting point! > > g. > > On Jul 28, 2008, at 4:32 PM, jim bower wrote: > > > Ross > > > > Great - thanks for support from an unexpected source. ;-). > > > > While it is completely contrary to the intuition of most > > experimentalists who are fond of their single neuron data, it is > > quite likely that individual neurons in at least the mammalian brain > > don't matter. Although, as I have mentioned before, one must be > > aware that neurobiologists pick stimuli to maximize the presumption > > that they do. > > > > On the other hand, in the extreme one could find oneself arguing a > > la Lashley that there is no structure in cortex and everything is > > completely distributed. This is also clearly not true. I have > > suspected for some time that the need for communication (as > > reflected very loosly in cortical oscillations) is what detemines > > the size of the population of neurons across which different types > > of computations are implemented but within which, individuals > > actually don't matter much. These populations are probably analogous > > to cortical areas. > > > > One other point however, is that abstract modelers tend to think of > > cortex as one thing and so do many neurobiologists who like to think > > in terms of things like the cortical microcircuit (or as in the case > > recently of the blue brain project of a cortical column, which > > actually doesn't exist - but anyway) in fact it is likely that > > cerebral cortex is a sequence of things. Getting at what cortex does > > will require that we understand where this sequence starts and ends. > > And, no time here and on my poor blackberry to discuss this, but if > > you are trying to sort these things out in the context of the visual > > system, as most are, you may be starting at the wrong end. It seems > > to me likely that it is the olfactory system that invented basic > > cortical circuitry. And there is a rather interesting story emerging > > there about just how random cortical activity and connectivity is. > > > > Jim bower > > > > Jim > > Sent via BlackBerry by AT&T > > > > -----Original Message----- > > From: "Ross Gayler" > > > > Date: Tue, 29 Jul 2008 08:07:20 > > To: > > Cc: > > Subject: Re: [Comp-neuro] Review announcement > > > > > > I had avoided weighing into this discussion on the grounds that my > > interest > > lies in abstract connectionist models that are quite remote from the > > physiological detail that has been the focus here. However, I would > > like to > > reinforce a comment that Jim Bower made. > > > >> the more sophisticated a coding system, the harder it is likely to > >> be to > > distinguish signal from "noise". > > > > Vector Symbolic Architectures are a family of abstract connectionist > > models > > that use very high-dimensional vectors to represent complex data > > structures > > that might be required for cognition. (Here's a quick challenge > > question for > > you. How might the brain represent "The astronomer believes that the > > little > > star orbits the big star"?) In VSA these data structures are > > represented by > > the pattern of activity over, say, 10,000 neurons (abstracted as a > > 10,000 > > dimensional vector). This is a thoroughly distributed > > representation in > > which there is no necessary significance to the activity of any > > individual > > neuron - it is only the overall pattern of activity that represents. > > > > Furthermore, and this is the point I wanted to make, individual > > patterns of > > activity are generally indistinguishable from random vectors. It is > > only > > the relationship between representing vectors that carries useful > > information. An external observer cannot decode the representing > > vectors > > without knowing what other vectors are stored in the clean-up memory > > of the > > system. I take these representations to be a very clear > > demonstration of > > Jim's point that in a sophisticated coding system it is harder to > > distinguish signal from noise. > > > > This paper (http://cogprints.org/3983/) provides some pointers into > > the VSA > > literature. The most extensive treatment of a specific > > instantiation of VSA > > is Tony Plate's book > > (http://csli-publications.stanford.edu/site/1575864304.html) > > > > > > Ross Gayler > > > > > > > > > > -----Original Message----- > > From: comp-neuro-bounces@neuroinf.org > > [mailto:comp-neuro-bounces@neuroinf.org] On Behalf Of jim bower > > Sent: Saturday, 26 July 2008 7:31 AM > > To: rod rinkus; comp-neuro-bounces@neuroinf.org; minai_ali@yahoo.com > > Cc: comp-neuro@neuroinf.org > > Subject: [Possible Spam] Re: [Comp-neuro] Review announcement > > > > One last plea for caution. No matter how convenient "noise" may be for > > abstract models, one can argue I think pretty convincingly that life > > itself > > differentiates itself from a-biological chemical evolution through > > its use > > of extreme degrees of structure and context specific organization to > > "beat" > > thermodynamics. > > > > As such anything that smells gaussian, disordered, something you can > > average > > over, or something that is completely generic rather than highly > > specific to > > the circumstances at hand (behavioral or computational) should raise > > concern. > > > > The long history of biology is that the closer you look, the more > > structure, > > not less structure you see. And everytime I have heard it suggested > > otherwise, closer inspection has suggested that we simply weren't > > giving > > biology enough credit. > > > > One quick example, a number of years ago, in fact during the first > > summer of > > the first computational neuroscience course in woods hole, a well know > > investigator of aplesia neurons gave a general MBL talk in which he > > sought > > to identify every conductance in the neuron of interest. In addition > > to the > > big "important" ones he found several small ones, whose presence he > > chalked > > up to sloppyness in protein transcription. > > > > Given my predispositions, I suggested that he change the temperature > > of the > > bath as aplesia live in tide pools. The next summer he gave a talk > > on the > > remarkable resiliance of neuronal function under different ambient > > temperatures confered by the complex conductances present in the cell > > membrane. > > > > While convenient, many of these "noise" mechanisms strike me as > > being simply > > not sophisticated enough. And I will point out now for the last > > time, that > > the more sophisticated a coding system, the harder it is likely to > > be to > > distinguish signal from "noise". > > > > Jim Bower > > Sent via BlackBerry by AT&T > > > > -----Original Message----- > > From: "rod rinkus" > > > > Date: Thu, 24 Jul 2008 00:33:59 > > To: > > Cc: > > Subject: RE: [Comp-neuro] Review announcement > > > > > > _______________________________________________ > > Comp-neuro mailing list > > Comp-neuro@neuroinf.org > > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > > > > _______________________________________________ > > Comp-neuro mailing list > > Comp-neuro@neuroinf.org > > http://www.neuroinf.org/mailman/listinfo/comp-neuro > > > Gary Cottrell 858-534-6640 FAX: 858-534-7029 > Computer Science and Engineering 0404 > IF USING FED EX INCLUDE THE FOLLOWING LINE: > CSE Building, Room 4130 > University of California San > Diego - > 9500 Gilman Drive # 0404 > La Jolla, Ca. 92093-0404 > > "Only connect!" -E.M. Forster > > "I am awaiting the day when people remember the fact that discovery > does not work by deciding what you want and then discovering it." > -David Mermin > > > Email: gary@ucsd.edu > Home page: http://www-cse.ucsd.edu/~gary/ > > > _________________________________________________________________ Confira v?deos com not?cias do NY Times, gols direto do Lance, videocassetadas e muito mais no MSN Video! http://video.msn.com/?mkt=pt-br -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080731/f34ed8f4/attachment-0001.html From malcolmdean at gmail.com Thu Jul 31 21:41:37 2008 From: malcolmdean at gmail.com (Malcolm Dean) Date: Fri Aug 1 13:09:59 2008 Subject: [Comp-neuro] Poulet 2008 [ Brain state regulates membrane potential synchrony; resting state; signal-to-noise ] In-Reply-To: <417b04640807311239t4ee7de54rd200fec1985638fa@mail.gmail.com> References: <417b04640807311239t4ee7de54rd200fec1985638fa@mail.gmail.com> Message-ID: <417b04640807311241x28a8c6d9m46b0a47edc4414b9@mail.gmail.com> "...a change in brain state dynamically and profoundly regulates cortical membrane potential synchrony during behaviour. The change in brain state is regulated by an internally generated signal... Active cortical brain states may therefore serve to augment the total cortical information processing capacity through decorrelation of membrane potential synchrony while increasing signal-to-noise ratios for AP initiation." http://www.nature.com/nature/journal/vaop/ncurrent/abs/nature07150.html Nature advance online publication 16 July 2008 doi:10.1038/nature07150 Internal brain state regulates membrane potential synchrony in barrel cortex of behaving mice James F. A. Poulet & Carl C. H. Petersen Internal brain states form key determinants for sensory perception, sensorimotor coordination and learning1, 2. A prominent reflection of different brain states in the mammalian central nervous system is the presence of distinct patterns of cortical synchrony, as revealed by extracellular recordings of the electroencephalogram, local field potential and action potentials. Such temporal correlations of cortical activity are thought to be fundamental mechanisms of neuronal computation3, 4, 5, 6, 7, 8, 9, 10, 11. However, it is unknown how cortical synchrony is reflected in the intracellular membrane potential (V m) dynamics of behaving animals. Here we show, using dual whole-cell recordings from layer 2/3 primary somatosensory barrel cortex in behaving mice, that the V m of nearby neurons is highly correlated during quiet wakefulness. However, when the mouse is whisking, an internally generated state change reduces the V m correlation, resulting in a desynchronized local field potential and electroencephalogram. Action potential activity was sparse during both quiet wakefulness and active whisking. Single action potentials were driven by a large, brief and specific excitatory input that was not present in the V m of neighbouring cells. Action potential initiation occurs with a higher signal-to-noise ratio during active whisking than during quiet periods. Therefore, we show that an internal brain state dynamically regulates cortical membrane potential synchrony during behaviour and defines different modes of cortical processing. -------------- next part -------------- An HTML attachment was scrubbed... URL: http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080731/adca7fcb/attachment.html From bower at uthscsa.edu Thu Jul 31 22:37:28 2008 From: bower at uthscsa.edu (jim bower) Date: Fri Aug 1 13:10:02 2008 Subject: [Comp-neuro] From Socrates to Ptolemy Message-ID: <214712052-1217536713-cardhu_decombobulator_blackberry.rim.net-880687806-@bxe111.bisx.prod.on.blackberry> Thanks to everyone, those who posted on comp neuro, and those who have responded to me directly, for your willingness to engage in this rather free wheeling discussion. Several have said and I agree "like the old days". I understand that the CNS meeting a couple of weeks ago in Portland also bore more of a resemblance to CNS meetings of old. All good news as one might otherwise be tempted to conclude that the big questions had already been answered. Two weeks ago I gave the introductory talk at the Latin American School for Computational Neuroscience (LASCON), which is the latest extension to the original course in computational neuroscience Christof Koch and I started 20 years ago in Woods Hole. In my introduction I warned about the tyranny of ideas and also made the point that the field was not necessarily as settled down as it might otherwise appear. I also told the students not to shy from asking basic questions (like is there really noise in the nervous system) just because it seems to be an accepted fact. So I wanted to thank you collectively for helping me make this point "live and in real time". ;-). The students watched with enthusiasm. Of course in my talk I also promoted the importance of realistic modeling in moving forward. On that note and to demonstrate the point, I asked them a simple question "what makes a good model". As usual the answers they gave (not more complex than it needs to be, able to replicate the observed data, ability to make measurable predictions, well matched to the analysis and modeling tools of the day, as easy to understand as possible, etc) would make Ptolemy's model of planetary motion a hands down choice over any other. Problem is, with Ptolemy there is no chance to learn something you didn't already know and no chance in particular to discover new underlying principles or structures, as these were already built into the structure of the model itself. As with Newton, Kepler, etc, the best protection is to first make your models realistic (even if it means you need to work with nasty math or invent new mathematical techniques, or don't really understand what is going on) and see if something new drops from the sky (or the tree, as the case may be. ;-). ). Again thanks to all, and especially the current moderators for tollerating this deviation from business as usual. And tchau from Brasil. Jim Sent via BlackBerry by AT&T From pprodrigues at liaad.up.pt Thu Jul 31 16:41:39 2008 From: pprodrigues at liaad.up.pt (Pedro Pereira Rodrigues) Date: Fri Aug 1 13:12:21 2008 Subject: [Comp-neuro] CFP - Two Weeks for Deadline - ACM SAC 2009 - Data Streams Track Message-ID: <4891CF23.2090305@liaad.up.pt> *** Apologies for cross-posting *** ACM SAC 2009 - TWO WEEKS FOR DEADLINE! ACM Symposium on Applied Computing The 24nd Annual ACM Symposium on Applied Computing Hilton Hawaiian Village Beach Resort & Spa Waikiki Beach, Honolulu, Hawaii, USA March 8 - 12, 2009 Data Streams Track http://www.liaad.up.pt/~jgama/SAC09/ IMPORTANT DATES Aug 16, 2008: Submission of papers Oct 11, 2008: Notification of acceptance/rejection Oct 25, 2008: Camera-ready copies of accepted papers DATA STREAMS TRACK - CALL FOR PAPERS The rapid development in information science and technology in general and in growth complexity and volume of data in particular have introduced new challenges for the research community. Many sources produce data continuously. Examples include sensor networks, wireless networks, radio frequency identification (RFID), customer click streams, telephone records, multimedia data, scientific data, sets of retail chain transactions, etc. These sources are called data streams. A data stream is an ordered sequence of instances that can be read only once or a small number of times using limited computing and storage capabilities. These sources of data are characterized by being open-ended, flowing at high-speed, and generated by non stationary distributions. TOPICS OF INTEREST We are looking for all possible contributions related to algorithms on data streams. Topics include (but are not restricted) to: Data Stream Models Data Stream Management Systems Data Stream Query Languages Continuous queries and Summarization from Data Streams Sampling Data Streams Single-Pass Algorithms Scalable Algorithms Change Detection Algorithms Clustering on Data Streams Classification and Regression on Data Streams Association Rules on Data Streams Feature Selection on Data Streams Visualization Techniques for Data Streams Evaluation of Data Streams Models Data Stream applications Sensor Networks Real-Time Applications PAPER SUBMISSION GUIDELINES Papers should be submitted in PDF using the SAC 2009 conference management system: http://sac.cs.iupui.edu/sac2009/ The author(s) name(s) and address(es) must NOT appear in the body of the paper, and self-reference should be in the third person. This is to facilitate blind review. Only the title should be shown at the first page without the author's information. The conference proceedings will be published by ACM. Hence, all accepted papers should be submitted in ACM 2-column camera ready format for publication in the symposium proceedings. The maximum number of pages allowed for the final papers is 5 pages (about 4000 words), with the option (at additional expense) to add up to three (3) more pages. There is a set of templates to support the required paper format for a number of document preparation systems at http://www.acm.org/sigs/pubs/proceed/template.html Each submitted paper will be fully refereed and undergo a blind review process by at least three referees. PROGRAM COMMITTEE Jose Avila, University Malaga, Spain Andre Carvalho, University S. Paulo, Brazil Antoine Cornu?jols, Institut National Paris, France Alfredo Cuzzocrea, University of Calabria, Italy Mohamed Gaber, Monash University, Australia Jo?o Gama, University Porto, Portugal Ricard Gavald?, Polytecnic Cataluna, Spain Georges H?brail, Telecom Paris, France Geoff Holmes, University Waikato, New Zealand Eamonn Keogh, University California, United States Ralf Klinkenberg, Rapid-I GmbH, Germany Miroslav Kubat, University Miami, United States Mark Last, University Ben Gorion, Israel Rosa Meo, University of Torino, Italy Pedro Rodrigues, University Porto, Portugal Josep Roure, Polytechnic Cataluna, Spain Elaine Sousa, University S. Paulo, Brazil Eduardo Spinosa, University S. Paulo, Brazil Min Wang, IBM, United States Sean Wang, University Vermon, United States Jiong Yang, Case Western Reserve University, United States Ying Yang, Australian Taxation Office, Australia Philip S. Yu, IBM, United States -------------- next part -------------- A non-text attachment was scrubbed... Name: pprodrigues.vcf Type: text/x-vcard Size: 335 bytes Desc: not available Url : http://www.neuroinf.org/pipermail/comp-neuro/attachments/20080731/87716219/pprodrigues-0001.vcf