From oreilly at psych.colorado.edu Thu Jun 3 11:41:58 2004 From: oreilly at psych.colorado.edu (Randall C. O'Reilly) Date: Thu, 3 Jun 2004 09:41:58 -0600 Subject: Postdoctoral position in cognitive control and language production Message-ID: <200406030941.58551.oreilly@psych.colorado.edu> Postdoctoral position in cognitive control and language production: Researchers collaborating through an NIMH-funded Interdisciplinary Behavioral Science Center are investigating the role of prefrontal cortex and mechanisms of cognitive control in language production. The research group includes Randy O'Reilly (University of Colorado, Boulder), Jonathan Cohen (Princeton University), Maryellen MacDonald (University of Wisconsin), and Todd Braver (Washington University). The postdoctoral position will be based at either Colorado or Princeton but will involve interaction with investigators at all four sites. The postdoctoral researcher will collaborate on computational and empirical studies of serial ordering and maintaining activation of information during utterance planning. Applicants should have a strong background in language production and computational modeling. Salary will be based on NIH postdoctoral scale, and the position may be renewable beyond the first year. The Ph.D. must be completed before beginning the position. Applicants should send a CV, a statement of research interests, representative preprints/reprints, and arrange for three letters of reference to be sent via email to Randall O'Reilly: oreilly at psych.colorado.edu (PDF format preferred). Applications will be reviewed on an ongoing basis, and will continue until the position is filled. The start date is flexible, but preferably within the coming academic year. For more information and contact addresses, see http://psych.colorado.edu/~oreilly/postdoc_ibsc.html From michael.spratling at kcl.ac.uk Thu Jun 3 13:44:35 2004 From: michael.spratling at kcl.ac.uk (Michael Spratling) Date: Thu, 03 Jun 2004 18:44:35 +0100 Subject: postdoc position, King's College London Message-ID: <40BF6383.8060206@kcl.ac.uk> An enthusiastic and well-qualified post-doctoral researcher is required to develop a biologically inspired neural network model that will be used to explore the effects of inter-regional cortical interactions in visual information processing. The project aims to understand the neural mechanisms underlying cortical region interactions and particularly the effects of top-down information on visual perception. It is hoped that this research will advance understanding of a number of perceptual processes (in areas such as attention, image segmentation, perceptual learning, and categorisation), and in so doing, identify computational principles appropriate for building improved machine vision systems. This project is part of an EPSRC-funded collaboration between the Division of Engineering at King's College London and the Centre for Brain and Cognitive Development (CBCD), Birkbeck College, London. The post-holder will be based in King's College, but will be expected to collaborate closely with researchers in the CBCD, where the predictions of the model will be empirically tested. Applicants should have a proven ability to carry out high quality research, have a genuine interest in the neural mechanisms underlying visual perception, and be keen to develop simulations of both psychological and neuro-physiological data. The successful applicant is expected to have a PhD in a relevant area, have a good knowledge of neural networks and/or computational neuroscience, be comfortable programming in C or C++ in a UNIX environment and ideally should be familiar with MATLAB. Informal inquiries can be directed to Dr Michael Spratling via e-mail: michael.spratling at kcl.ac.uk. The position is available from the 1st October 2004 for a period of 33 months. The starting salary is at SP6 on the RA1A scale, currently =A323,144 per annum inclusive of London Allowance. Further particulars are available by contacting the Personnel Office, King's College London, Strand, London WC2R 2LS, strand-recruitment at kcl.ac.uk, fax 020 7848 1352. Please quote reference W1/CEM/46/04 on all correspondence. The closing date for receipt of completed application forms is: 25 June 2004. Equality of opportunity is College policy. From seth at nsi.edu Mon Jun 7 19:00:41 2004 From: seth at nsi.edu (Anil K Seth) Date: Mon, 07 Jun 2004 16:00:41 -0700 Subject: Workshop on Neurorobotics and Neuroinformatics Message-ID: <5.1.0.14.0.20040607155951.049d4c20@mail.nsi.edu> We are pleased to announce the following workshop, to be held on July 17, 2004 in Santa Monica, CA, as part of the 8th International Conference on the Simulation of Adaptive Behavior (SAB2004): Neurorobotic Models in Neuroscience and Neuroinformatics Adaptive behavior in biological organisms results from interactions among brains, bodies, and environments. We propose in this workshop to explore neurorobotic approaches to understanding these interactions. A key feature of these approaches is the incorporation of aspects of neuroanatomy or neurophysiology that allow comparison with empirical data. More information is available at www.nsi.edu/users/seth/SABwsp.htm We have a great lineup of invited speakers, including: Hiroshi Kimura, Stefan Schaal, Auke Ijspeert, Angelo Arleo, Philippe Gaussier, Jeff Krichmar, Kenji Doya, Ricardo Chavarriaga, Will Alexander, Olaf Sporns, and Steve Potter. We also have a great lineup of coffee breaks, allowing plenty of time for informal discussion. The day will end with dinner somewhere in Santa Monica. We expect this workshop to have a significant impact in both Neuroinformatics (there will be a special issue of the Neuroinformatics journal based on the workshop contents) and in Adaptive Behavior. Come and be a part of it! You can register for only $90 (dinner not included) via the main SAB website: www.isab.org.uk/sab2004/ Please feel free to contact me (seth at nsi.edu) with any questions. Best, Anil Seth Jeff Krichmar Olaf Sporns Auke Ijspeert Ricardo Chavarriaga ----------------------------------------------------------------------------- Anil K Seth, D.Phil., The Neurosciences Institute, 10640 John Jay Hopkins Drive, San Diego, CA 92121, tel: (858) 626 2066, fax: (858) 626 2099 email: seth at nsi.edu, web: www.nsi.edu/users/seth/ From schunn+ at pitt.edu Tue Jun 8 13:45:05 2004 From: schunn+ at pitt.edu (Christian Schunn) Date: Tue, 08 Jun 2004 13:45:05 -0400 Subject: ICCM2004 registration and poster submission info Message-ID: <93AEBAF2-B973-11D8-B937-000393BCD4AC@pitt.edu> International Conference on Cognitive Modeling Pittsburgh, USA Carnegie Mellon University + University of Pittsburgh July 29 - Aug 1, 2004 Online registration for ICCM2004 is now open. Early registration rates are valid until June 28, 2004. Follow the registration link at: http://simon.lrdc.pitt.edu/~iccm/ Early registration: Students $50, Nonstudents $100 Full registration (after June 28th): Students $75, Nonstudents $150. These fees include the cost of the program, the proceedings, the poster reception, the banquet, the break food, and a few other surprises (i.e., the conference fee is surprisingly low). Full and half-day tutorials (on CHREST, Cogent, EPIC, and SOAR) are taking place on the 29th of July, at an extra cost. ($40/half day for nonstudents, $30/half day for students). Registration for tutorials, parking, and staying on campus at the dorms are all done on the same registration page. Those choose to stay at local hotels will make their hotel reservations directly with the hotels. Links are available at: http://simon.lrdc.pitt.edu/~iccm/hotels.html Plenary speakers are Ken Forbus and Michael Mozer. Special symposia include modeling of eye-movements and the pokerbot competition results. The schedule of talks will appear later this week. The submission deadline for 2-page poster-abstracts for ICCM2004 is June 15, 23:59 EDST To submit a poster abstract, go to: http://simon.lrdc.pitt.edu/~iccm/abstract-submit.html Formatting instructions can be found at: http://simon.lrdc.pitt.edu/~iccm/ICCM2004-Submissions.html The schedule of talks will be posted later this week. For further information about ICCM2004, see http://simon.lrdc.pitt.edu/~iccm/ CONFERENCE CHAIRS Marsha Lovett (lovett at cmu.edu) Christian Schunn (schunn at pitt.edu) Christian Lebiere (clebiere at maad.com) Paul Munro (pmunro at mail.sis.pitt.edu) From wolpert at ion.ucl.ac.uk Mon Jun 14 04:44:49 2004 From: wolpert at ion.ucl.ac.uk (Daniel Wolpert) Date: Mon, 14 Jun 2004 09:44:49 +0100 Subject: Postdoctoral Position in Computational Sensorimotor Control Message-ID: A postdoctoral position in human sensorimotor control is available in the laboratory of Daniel Wolpert at the Institute of Neurology, University College London. Ongoing work in the laboratory examines Bayesian and optimal control processes in human motor learning, sensorimotor integration and control. The successful applicant will be expected to conduct independent research involving computationally motivated experimental studies in humans. Applicants should have a PhD and technical expertise and skills relevant to the study of human movement; however, applicants with a strong background in computational neuroscience who wish to learn experimental approaches will also be considered. Experience with Matlab & C++ would be very advantageous. University College London has a large concentration of researchers in the areas of motor control and computational neuroscience, including the neighbouring Gatsby Computational Neuroscience Unit and Institute of Cognitive Neuroscience. Further details of the post and laboratory facilities can be found on www.hera.ucl.ac.uk. Informal enquiries can be addressed by email to Professor Daniel Wolpert (wolpert at ion.ucl.ac.uk). The position is available from September 2004 for one year in the first instance. Starting salary up to =A333,849 pa inclusive, depending on experience, superannuable. Applicants should provide by 30 June 2004 (preferably by email to e.bertram at ion.ucl.ac.uk): - a 1 page statement of research interests - copy of CV (2 if sent by post) - names and contact details of 3 referees - 1 copy of Declaration (required - see further details of posts) - Equal Opportunities form (optional - see further details of posts) to: Miss E Bertram, Assistant Secretary (Personnel) Institute of Neurology Queen Square London WC1N 3BG Fax: +44 (0)20 7278 5069 Email: e.bertram at ion.ucl.ac.uk UCL Taking Action for Equality From l.wiskott at biologie.hu-berlin.de Mon Jun 14 13:15:35 2004 From: l.wiskott at biologie.hu-berlin.de (Laurenz Wiskott) Date: Mon, 14 Jun 2004 19:15:35 +0200 Subject: Postdoc position available in Berlin, Germany Message-ID: <16589.56631.558957.402173@huxley.biologie.hu-berlin.de> A postdoc position will be available January 2005 in the group of Laurenz Wiskott Institute for Theoretical Biology Humboldt-University Berlin for modeling hippocampal function including adult neurogenesis more info at http://itb.biologie.hu-berlin.de/~wiskott/jobs2.html From arno at salk.edu Mon Jun 14 13:37:58 2004 From: arno at salk.edu (Arnaud Delorme) Date: Mon, 14 Jun 2004 10:37:58 -0700 Subject: First EEGLAB Workshop Message-ID: <40CDE276.6090003@salk.edu> Call for pre-registration for the ... First EEGLAB Workshop EEGLAB is a growing open source Matlab environment for signal processing and visualization of EEG, MEG and other electrophysiological signals. For complete information, visit http://sccn.ucsd.edu/eeglab The first three-day EEGLAB workshop will be held Oct. 28-30, 2004, following the Society for Neuroscience meeting in San Diego, at the San Diego Supercomputer Center of the University of California San Diego in La Jolla. Its goals are to introduce EEGLAB to those considering its use in their research and to help current users perform more advanced analyses. Pre-registration applications are being accepted now through July 20. For detailed information and application, visit http://sccn.ucsd.edu/eeglab/workshop/ The workshop faculty, Arnaud Delorme Julie Onton Tzyy-Ping Jung Scott Makeig -- *Arnaud Delorme, Ph.D.* Computational Neurobiology Lab, Salk Institute 10010 North Torrey Pines Road La Jolla, CA 92037 USA *Tel* : /(+1)-858-458-1927 ext 15/ *Fax* : /(+1)-858-458-1847/ *Web page *: www.sccn.ucsd.edu/~arno *To think upon*: Life is either a daring adventure or nothing. /Helen Keller/ From bassis at dsi.unimi.it Tue Jun 15 11:04:33 2004 From: bassis at dsi.unimi.it (Simone Bassis) Date: Tue, 15 Jun 2004 17:04:33 +0200 Subject: WIRN 04 DEADLINE EXTENSION Message-ID: <200406151504.31592.bassis@dsi.unimi.it> [we apologize for multiple receipts] Due to numerous solicitations the deadline for submitting papers to WIRN04 (XV Italian Workshop on Neural Networks, http://siren.dsi.unimi.it/conferences/SI/html) has been delayed to June 26. The electronic submission is regularly open (a temporary inconvenience has been removed). Waiting for you in Perugia, best regards Bruno Apolloni From dgw at MIT.EDU Tue Jun 15 14:27:11 2004 From: dgw at MIT.EDU (David Weininger) Date: Tue, 15 Jun 2004 14:27:11 -0400 Subject: book announcement--Forbes Message-ID: <200406151427116322@outgoing.mit.edu> I thought readers of the Connectionists List might be interested in this book. For more information, please visit http://mitpress.mit.edu/0262062410/ Thank you! Best, David Imitation of Life How Biology Is Inspiring Computing Nancy Forbes As computers and the tasks they perform become increasingly complex, researchers are looking to nature--as model and as metaphor--for inspiration. The organization and behavior of biological organisms present scientists with an invitation to reinvent computing for the complex tasks of the future. In Imitation of Life Nancy Forbes surveys the emerging field of biologically inspired computing, looking at some of the most impressive and influential examples of this fertile synergy. Forbes points out that the influence of biology on computing goes back to the early days of computer science--John von Neumann, the architect of the first digital computer, used the human brain as the model for his design. Inspired by von Neumann and other early visionaries, as well as by her work on the "Ultrascale Computing" project at the Defense Advanced Research Projects Agency (DARPA), Forbes describes the exciting potential of these revolutionary new technologies. She identifies three strains of biologically inspired computing: the use of biology as a metaphor or inspiration for the development of algorithms; the construction of information processing systems that use biological materials or are modeled on biological processes, or both; and the effort to understand how biological organisms "compute," or process information. Forbes then shows us how current researchers are using these approaches. In successive chapters, she looks at artificial neural networks; evolutionary and genetic algorithms, which search for the "fittest" among a generation of solutions; cellular automata; artificial life--not just a simulation, but "alive" in the internal ecosystem of the computer; DNA computation, which uses the encoding capability of DNA to devise algorithms; self-assembly and its potential use in nanotechnology; amorphous computing, modeled on the kind of cooperation seen in a colony of cells or a swarm of bees; computer immune systems; bio-hardware and how bioelectronics compares to silicon; and the "computational" properties of cells. Nancy Forbes works as a science and technology analyst for the federal government. She has advanced degrees both in physics and the humanities, and has served as a contributing editor for The Industrial Physicist and Computing in Science and Engineering. 6 x 9, 176 pp., 48 illus., cloth, ISBN 0-262-06241-0, $25.95 ______________________ David Weininger Associate Publicist The MIT Press 5 Cambridge Center, 4th Floor Cambridge, MA 02142 617 253 2079 617 253 1709 fax http://mitpress.mit.edu From l.s.smith at cs.stir.ac.uk Wed Jun 16 04:45:17 2004 From: l.s.smith at cs.stir.ac.uk (Professor Leslie Smith) Date: Wed, 16 Jun 2004 09:45:17 +0100 Subject: BICS2004, Stirling, Scotland: Call for Participation Message-ID: <40D0089D.3080203@cs.stir.ac.uk> Brain Inspired Cognitive Systems - BICS2004: call for participation. University of Stirling, Scotland, UK August 29 - September 1, 2004 First International ICSC Symposium on Cognitive Neuro Science Chair: Prof. Igor Aleksander, Imperial College London, U.K Second International ICSC Symposium on Biologically Inspired Systems Chair: Prof. Leslie Smith, University of Stirling, U.K. Third International ICSC Symposium on Neural Computation Chair: Dr. Amir Hussain, University of Stirling, U.K. Plenary Speakers: Prof. Rodney Douglas Prof. Graham Hesketh Professor Erkki Oja Owen Holland Prof JG Taylor Prof David Willshaw BICS2004 is being held in Stirling, a historic city located in the heart of Scotland. The conference venue and accommodation are all on the University campus, which is convenient for visiting the ancient castle and Wallace monument, or for going further afield and exploring the beautiful countryside. Register now for this conference: see the website for the full programme and for registration: http://www.icsc-naiso.org/conferences/bics2004/bics-cfp.html -- Professor Leslie S. Smith, Dept of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA, Scotland l.s.smith at cs.stir.ac.uk Tel (44) 1786 467435 Fax (44) 1786 464551 www http://www.cs.stir.ac.uk/~lss/ UKRI IEEE NNS Chapter Chair: http://www.cs.stir.ac.uk/ieee-nns-ukri/ From hinton at cs.toronto.edu Wed Jun 16 10:17:40 2004 From: hinton at cs.toronto.edu (Geoffrey Hinton) Date: Wed, 16 Jun 2004 10:17:40 -0400 Subject: new book on semantic cognition Message-ID: <04Jun16.101742edt.453135-627@jane.cs.toronto.edu> There is a nice new book showing that distributed representations in connectionist networks can explain a lot of psychological data: ___________________________________________ Semantic Cognition A Parallel Distributed Processing Approach Timothy T. Rogers and James L. McClelland http://mitpress.mit.edu/0262182394/ This groundbreaking monograph offers a mechanistic theory of the representation and use of semantic knowledge, integrating the strengths and overcoming many of the weaknesses of hierarchical, categorization-based approaches, similarity-based approaches, and the approach often called "theory theory." Building on earlier models by Geoffrey Hinton in the 1980s and David Rumelhart in the early 1990s, the authors propose that performance in semantic tasks arises through the propagation of graded signals in a system of interconnected processing units. The representations used in performing these tasks are patterns of activation across units, governed by weighted connections among them. Semantic knowledge is acquired through the gradual adjustment of the strengths of these connections in the course of day-to-day experience. The authors show how a simple computational model proposed by Rumelhart exhibits a progressive differentiation of conceptual knowledge, paralleling aspects of cognitive development seen in the work of Frank Keil and Jean Mandler. The authors extend the model to address aspects of conceptual knowledge acquisition in infancy, disintegration of conceptual knowledge in dementia, "basic-level" effects and their interaction with expertise, and many findings introduced to support the idea that semantic cognition is guided by naive, domain-specific theories. From nello at wald.ucdavis.edu Wed Jun 16 19:22:27 2004 From: nello at wald.ucdavis.edu (Nello Cristianini) Date: Wed, 16 Jun 2004 16:22:27 -0700 (PDT) Subject: Book: Kernel Methods for Pattern Analyis Message-ID: <20040616162002.I6178-100000@anson.ucdavis.edu> New book, available now: Kernel Methods for Pattern Analysis John Shawe-Taylor & Nello Cristianini Cambridge University Press, 2004 ISBN: 0521813972 http://www.kernel-methods.net/ Kernel methods provide an approach to pattern analysis that has developed over the last 12 years since the introduction of Support Vector Machines. This new book has two main aims: to provide an introduction to this approach and act as a cook-book for practitioners. The three parts, Concepts, Algorithms and Kernels, contain dozens of algorithms and kernels, pseudocode and detailed derivations, while Matlab code is freely available from the book website. The book provides a comprehensive account of the field of Kernel Methods for Pattern Analysis, summarizing more than a decade of expansion of these exciting and successful techniques. From nipsinfo at salk.edu Thu Jun 17 17:22:30 2004 From: nipsinfo at salk.edu (nipsinfo@salk.edu) Date: Thu, 17 Jun 2004 14:22:30 -0700 Subject: NIPS 2004 Survey Message-ID: To: NIPS Attendees and Connectionists: We are asking for a few minutes of your time to complete an online survey consisting of seven questions. The survey deals with a some major changes to the format of the 2004 Neural Information Processing Systems Conference currently under consideration. The impetus for these possible changes is to accommodate the growth in submissions in recent years, and the diverse demographics of Conference attendees. One concern that has been raised regarding the increased size of the meeting is the unwieldy scale of the Poster Sessions. Another is that while the representation in machine learning at NIPS is growing, many feel that its focus is becoming narrower; the result of this has been that many prominent computational neuroscientists, cognitive modelers, etc. have stopped attending NIPS because they believe the meeting to be overly dominated by machine learning. The survey can be found at the NIPS 2004 website: https://register.nips.salk.edu/surveys/survey.php?id=3 Your responses to these questions, as well as any additional input you may have, will be very valuable in shaping the future of NIPS, and we very much appreciate your participation. Neural Information Processing Systems Conference 2004 From bogus@does.not.exist.com Thu Jun 17 11:55:02 2004 From: bogus@does.not.exist.com () Date: Thu, 17 Jun 2004 16:55:02 +0100 Subject: 9th Neural Computation and Psychology Workshop: Extended deadline and Limited Travel grants Message-ID: <736F0925D69F9941B3BA8AEED0F5E75C029A746B@02-CSEXCH.uopnet.plymouth.ac.uk> From d.mareschal at bbk.ac.uk Thu Jun 17 09:29:19 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Thu, 17 Jun 2004 14:29:19 +0100 Subject: connectionism and development Message-ID: Dear all, The following special section in Developmental Science 7(2) may be of interest to readers of this list. SECIAL SECTION Beyond Backprogation: emerging trends in connectionist models of 131 development Matthew Schlesinger and Domenico Parisi Autoassociator networks: insights into infant cognition 133 Sylvain Sirois Hebbian learning and development 141 Yuko Munakata and Jason Pfaffly Modeling developmental transitions in adaptive resonance theory 149 Maartje E. J. Raijmakers and Peter C. M. Molenaar Evolving agents as a metaphor for the developing child 158 Matthew Schlesinger Commentary 165 David Klahr This issue also contains a scientific obituary for Elizabeth Bates ================================================= Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================= From arthur at tuebingen.mpg.de Fri Jun 18 21:43:55 2004 From: arthur at tuebingen.mpg.de (Arthur Gretton) Date: 19 Jun 2004 03:43:55 +0200 Subject: extended deadline for ICML/COLT session on kernel methods Message-ID: <1087609435.1030.5.camel@bertie> A non-text attachment was scrubbed... Name: not available Type: multipart/alternative Size: 919 bytes Desc: not available Url : https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/0200062004/17d9c47f/attachment.bin From special at math.unipd.it Fri Jun 18 05:17:04 2004 From: special at math.unipd.it (Alessandro Sperduti) Date: Fri, 18 Jun 2004 11:17:04 +0200 (CEST) Subject: Cfp: Special Issue Message-ID: ---------------------------------------------------------------------- C A L L F O R P A P E R S Special Issue of Neural Networks on Neural Networks and Kernel Methods for Structured Domains Co-editors:Barbara Hammer, Craig Saunders, Alessandro Sperduti (http://www.informatik.uni-osnabrueck.de/barbara/specialNN/) ---------------------------------------------------------------------- In recent years several researchers have started to consider the adaptive processing of structured data. This interest is motivated by two main reasons: i) several very important computational problems in Bioinformatics, Chemistry, document classification and filtering (just to name a few), require the use of some Machine Learning procedure to be properly treated because their complexity does not allow a formal and precise definition of the problem and thus no algorithmic solution to the problem is known; on the other hand experimental data is available and usable by a Machine Learning approach; ii) in many of the above problems, the objects of interest are more naturally represented via varying-size structured representations, such as sequences, strings, trees, directed or undirected graphs. Moreover, it is commonly believed that any computational process representing these objects via flat fixed-size vectorial representations, risks to discard structural information relevant for solving the task at hand. The aim of the special issue is to bring together recent works developed mainly into the fields of Recurrent and Recursive Neural Networks, and Kernel Methods, with the hope to function as a conceptual support for the improvement of current technologies and the growth of new integrated methodologies for learning in structured domains. TOPICS: Recurrent and Recursive Neural Networks, Self-Organizing Maps for Structures, Support Vector Machines and Kernel Methods for Structured Domains, Fisher Kernels, Probabilistic Models for Structured Domains, Applications of the above methods to Structured Domains, other relevant topics. ---------------------------------------------------------------------- I M P O R T A N T D A T E S Deadline for submission October 11, 2004 Notification of acceptance: March 14, 2005 Deadline for final paper: June 6, 2005 ---------------------------------------------------------------------- P A P E R S U B M I S S I O N Format: http://www.informatik.uni-osnabrueck.de/barbara/specialNN/ Papers must be submitted in electronic form (either Postscript or PDF) to the following e-mail address: specialNN.structure at math.unipd.it Hard-paper submissions are accepted only if the authors do not have access to electronic facilities, and should be mailed to the following address: Prof. Alessandro Sperduti Dipartimento di Matematica Pura ed Applicata Via Belzoni 7 35131 Padova ITALY ---------------------------------------------------------------------- From scheler at stanford.edu Sat Jun 19 22:26:15 2004 From: scheler at stanford.edu (Gabriele Scheler) Date: Sat, 19 Jun 2004 19:26:15 -0700 Subject: Papers on Neural Plasticity Message-ID: <1087698375.40d4f5c75704a@webmail.stanford.edu> Dear colleagues, Here is an announcement for two recent papers (and an earlier one from 2003) on non-LTP-like brain plasticity. available from arxiv.org Summary: ------------------------------ The evidence for learning and memory expressed by the distribution of receptors for other neurochemicals, prominently dopamine, acetylcholine etc. is reviewed. Theoretical models are outlined (a) for synaptic gating by presynaptic receptor activity and (b) for intrinsic, whole-neuron adaptivity in membrane excitation ('transfer') functions. The upshot is that learning and memory may very likely be expressed by the adjustment of G-protein coupled receptors, even though the amount of heterogeneity (storage capacity) and the amount of long-term persistence of adaptivity are not well investigated yet. Known to be the target of many psychoactive substances, and prominent in addiction, fear and social learning, we conjecture that neuromodulatory plasticity has a quite general function in 'emotional learning' in many parts of the brain. --------------------------------- Regulation of Neuromodulator Efficacy - Implications for Whole-Neuron and Synaptic Plasticity. Progress in Neurobiology 72(6), April 2004. http://www.arxiv.org/pdf/q-bio.NC/0403011 Memorization in a neural network with adjustable transfer function and conditional gating. online since March 2004. http://www.arxiv.org/pdf/q-bio.MN/0401039 Presynaptic modulation as fast synaptic switching: state-dependent modulation of task performance. Proceedings of IJCNN'03, Portland, Oregon, June 2003. http://www.arxiv.org/pdf/cs.NE/0401020 -------------------------------- Dr. G. Scheler Research Scientist ISLE/CSLI 25 Ventura Hall Stanford, Ca. 94305 http://www.stanford.edu/~scheler From auke.ijspeert at epfl.ch Tue Jun 22 03:57:11 2004 From: auke.ijspeert at epfl.ch (Auke Jan Ijspeert) Date: Tue, 22 Jun 2004 09:57:11 +0200 Subject: SAB2004: calls for participation and abstracts (ext'd deadline) Message-ID: <40D7E657.3070601@epfl.ch> Dear Connectionists, Researchers in adaptive behavior, neural computation, and artificial neural networks applied to robotics might be interested in the next SAB2004, From Animals to Animats 8 conference that will take place in Santa Monica July 13-17 2004. See the call for participation and the call for abstracts below. Note that the deadline for the Last-Minute-Results abstract submission has been extended until July 9th. Best regards, Auke Ijspeert -----SAB2004 Call for participation and abstracts (ext'd deadline)----- SAB2004, From Animals to Animats 8, The Eighth International Conference on the Simulation of Adaptive Behavior Santa Monica (Los Angeles), July 13-17 2004 http://www.isab.org/sab04/ ----- Call for participation ----- The objective of this interdisciplinary conference is to bring together researchers in computer science, artificial intelligence, alife, control, robotics, neurosciences, ethology, and related fields so as to further our understanding of the behaviors and underlying mechanisms that allow natural and artificial animals to adapt and survive in uncertain environments. The conference will focus on experiments with well-defined models --- robot models, computer simulation models, mathematical models --- designed to help characterize and compare various organizational principles or architectures underlying adaptive behavior in real animals and in synthetic agents, the animats. The preliminary technical program can be found at http://www.isab.org/sab04/program/ See http://www.isab.org/sab04/ for registration and more details. -----Call for abstracts for the Last-Minute-Results poster session----- SAB2004 introduces a new feature: the Last-Minute-Results poster session (July 9th, extended deadline). This special session will offer researchers in Adaptive Behavior the opportunity to present their most recent results at SAB2004 (http://www.isab.org/sab04/). The goal of this session is to provide an informal setting in which participants can reveal and discuss their latest results and developments at the time of the conference (13-17 July 2004). Researchers are invited to submit a two-page abstract describing recent results/developments. Abstracts should be sent electronically as a PDF file to the conference general email address, sab2004 at isab.org, with "Last Minute Results" in the email Subject line. The call for abstracts is open to all (i.e. to people both with and without an accepted paper at the conference). There is a limit of one abstract/poster per participant as a first author. After review by the conference chairs, authors of accepted abstracts will be allowed to present their results as a poster in a special Last-Minute-Results session. Note that the abstracts will not be considered as publications and will not be included in the conference proceedings. This also means that you keep the abstracts' copyrights. Deadlines: July 9th: Abstract submission deadline (but note that there is a limited number of slots, and that posters will be accepted after a short review on a first come, first served basis) July 13-17 2004: SAB'04 conference Contributions treating any of the following topics from the perspective of adaptive behavior will receive special emphasis: The Animat approach Characterization of agents and environments Passive and active perception Motor control Visually-guided behaviors Action selection Behavioral sequencing Navigation and mapping Internal models and representation Learning and development Motivation and emotion Collective and social behavior Emergent structures and behaviors Neural correlates of behavior Evolutionary and co-evolutionary approaches Autonomous robotics Humanoid robotics Software agents and virtual creatures Applied adaptive behavior Animats in education Philosophical and psychological issues X-Antivirus-NBU.BG-Mail-From: office at cogs.nbu.bg via anna-karenina X-Antivirus-NBU.BG: 1.22st (Clear:RC:1(192.168.50.47):. Processed in 0.028888 secs Process 29365) From office at cogs.nbu.bg Wed Jun 23 13:26:59 2004 From: office at cogs.nbu.bg (Cogs Office) Date: Wed, 23 Jun 2004 10:26:59 -0700 Subject: Cognitive Science, NBU - Announcement Message-ID: THE CENTRAL AND EAST EUROPEAN CENTER FOR COGNITIVE SCIENCE INVITES APPLICATIONS FOR: 1. One-year post-doctoral fellowships for the 2003/2004 academic year (with a possible extension); 2. Two-year Ph.D. studentships (starting in October 2004). For more information: http://www.nbu.bg/cogs/events/2004/positions.html Central and East European Center for Cognitive Science New Bulgarian University 21 Montevideo Str. Sofia 1618, Bulgaria phone: (+3592) 8110-401 e-mail: office at cogs.nbu.bg From nips04pub at hotmail.com Thu Jun 24 16:25:04 2004 From: nips04pub at hotmail.com (John Platt) Date: Thu, 24 Jun 2004 13:25:04 -0700 Subject: NIPS 2004 Call for Demonstrations Message-ID: CALL FOR DEMONSTRATIONS Neural Information Processing Systems --- Natural and Synthetic Monday, December 13 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc DEADLINE FOR DEMONSTRATION PROPOSALS: August 1, 2004 Based on their success in 2002 and 2003, the Neural Information Processing Systems Conference will again include a separate track for Demonstrations. The Demonstrations will take place in parallel with the NIPS poster sessions on Tuesday and Wednesday evenings (December 14 and 15). The key requirements for Demonstrations is that they be LIVE and INTERACTIVE and that they present a compelling view of an emerging technology. Past Demonstrations have covered a very wide range. Areas of interest for the Demonstrations track have previously included the following: analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, robotics, bioMEMS (microelectromechanical systems), biomedical instrumentation, neural prostheses, photonics, real-time multimedia systems, large-scale neural emulators, software Demonstrations of novel algorithms, and open-source software toolboxes. NIPS is a continually-evolving interdisciplinary Conference, which attracts cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The Demonstration track enables researchers to highlight scientific advances, systems, and technologies in ways that go beyond conventional poster presentations. It will provide a unique forum for demonstrating advanced technologies (hardware and software), and fostering the direct exchange of knowledge. We hope that this track will stimulate interactions between researchers from different fields or approaches. Submissions accepted in the Demonstrations track will be published on the NIPS web site, but will not appear in printed proceedings. However, submitting your work to the Demonstration track does not preclude the submission of a companion paper to the regular NIPS Conference; joint submissions are very much encouraged. We also encourage authors submitting Demonstrations to consider organizing a Workshop at NIPS 2004. There will be a separate room for these Demonstrations and participants will have access to power strips, tables and poster boards. Monitors will also be provided on request at their rental cost. Participants are responsible for ensuring that their Demonstration is sufficiently portable; additional hardware beyond that specified above might be provided at cost, if readily available. SUBMISSION INSTRUCTIONS: Proposals for Demonstrations will be reviewed by the Demonstrations Co-Chairs. Demonstration proposals should be submitted via the web form available at http://nips.cc. Demonstrators will be asked to enter information about the nature of the Demonstration, in particular they will be asked to describe first the user experience and then the underlying technology. Proposals that are simply papers in disguise will be rejected, this session is for live, interactive experiences that compellingly demonstrate new technology. The Demonstration track is not an alternative poster session. Proposers will also be asked about the present state of their Demonstration in order that the co-chairs may judge whether the Demonstration can actually be made functional. Past experience has shown that simpler Demonstrations that make one point are usually more interesting to attendees. Complex Demonstrations involving multiple technologies and partners have not been as effective. NIPS 2004 DEMONSTRATIONS CO-CHAIRS: Tobi Delbruck, Institute of Neuroinformatics, University of Zurich, ETH Zurich; Timmer Horiuchi, Electrical and Computer Engineering Department, Institute for Systems Research and Neuroscience and Cognitive Science Program, University of Maryland. DEADLINE FOR DEMONSTRATION PROPOSALS: August 1, 2004 From mjhealy at chango.eece.unm.edu Fri Jun 25 21:02:48 2004 From: mjhealy at chango.eece.unm.edu (M. Healy) Date: Fri, 25 Jun 2004 19:02:48 -0600 (MDT) Subject: Technical report available Message-ID: A technical report on a mathematical model of the semantics of neural networks is now available on the Dspace archive in the collection 2004. To access it, go to the University of New Mexico site at https://repository.ece.unm.edu/dspace/ and search on http://hdl.handle.net/1928/33 . Here's an abstract: Neural Networks, Knowledge, and Cognition: A Mathematical Semantic Model Based upon Category Theory, by M. J. Healy and T. P. Caudell Category theory can be applied to mathematically model the semantics of cognitive neural systems. We discuss semantics as a hierarchy of concepts, or symbolic descriptions of items sensed and represented in the connection weights distributed throughout a neural network. The hierarchy expresses subconcept relationships, and in a neural network it becomes represented incrementally through a Hebbian-like learning process. The categorical semantic model described here explains the learning process as the derivation of colimits and limits in a concept category. It explains the representation of the concept hierarchy in a neural network at each stage of learning as a system of functors and natural transformations, expressing knowledge coherence across the regions of a multi-regional network equipped with multiple sensors. The model yields design principles that constrain neural network designs capable of the most important aspects of cognitive behavior. Please let me know right away if there are any problems. I'll warmly receive any questions or comments. - Mike From terry at salk.edu Sat Jun 26 12:32:25 2004 From: terry at salk.edu (Terry Sejnowski) Date: Sat, 26 Jun 2004 09:32:25 -0700 (PDT) Subject: NEURAL COMPUTATION 16:8 In-Reply-To: <200405242248.i4OMmXL28618@dax.snl.salk.edu> Message-ID: <200406261632.i5QGWPO06371@dax.salk.edu> Neural Computation - Contents - Volume 16, Number 8 - August 1, 2004 LETTERS A Coarse-to-Fine Disparity Energy Model with Both Phase-Shift and Position-Shift Receptive Field Mechanisms Yuzhi Chen and Ning Qian A Preference for Phase-Based Disparity in a Neuromorphic Implementation of the Binocular Energy Model Eric K.C. Tsang and Bertram E. Shi Learning Classification in the Olfactory System of Insects Ramon Huerta, Thomas Nowotny, Marta Garcia-Sanchez, H.D.I Abarbanel, and M.I. Rabinovich Adaptive Blind Separation with Unknown Number of Sources Ji-Min Ye, Xiao-Long Zhu, and Xian-Da Zhang Unsupervised Spike Detection and Sorting with Wavelets and Superparamagnetic Clustering R. Quian Quiroga, Z. Nadasdy and Y. Ben-Shaul Decomposition Methods for Linear Support Vector Machines Wei-Chun Kao, Kai-Min Chung, Chia-Liang Sun, and Chih-Jen Lin An Asymptotic Statistical Theory of Polynomial Kernel Methods Kazushi Ikeda A Neural Root Finder of Polynomials Based on Root Moments De-Shuang Huang, Horace H. S. Ip, and Zheru Chi ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2004 - VOLUME 16 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $108 $54 $57.78 Individual $95 $101.65 $143 $85 $90.95 Institution $635 $679.45 $689 $572 $612.04 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From pmr at di.fc.ul.pt Tue Jun 29 04:21:37 2004 From: pmr at di.fc.ul.pt (pmr@di.fc.ul.pt) Date: Tue, 29 Jun 2004 09:21:37 +0100 Subject: Extension to Delta Rule already studied before? Message-ID: <1088497297.40e1269148dc5@webservices.di.fc.ul.pt> Dear connectionists, while working on my Ph.D. thesis, I've been using neural nets that apply the 'Delta Rule'. Nevertheless, for some instances of the problems I've been addressing, the 'Delta Rule' does not seem appropriate: it promotes weights that do not differ substantially. While trying to overcome this property, I've came across the following rule: w_ij = w_ij + a_i * (t_j - a_j) * LR * w_ij where... w_ij - weight of the connection from input neuron i to output neuron j a_i - activity of input neuron i t_j - target activity for output unit j a_j - activity of output neuron j LR - learning rate Has this learning rule been studied before? Are there references ? Many thanks for your attention, kind regards, pedro From oreilly at psych.colorado.edu Thu Jun 3 11:41:58 2004 From: oreilly at psych.colorado.edu (Randall C. O'Reilly) Date: Thu, 3 Jun 2004 09:41:58 -0600 Subject: Postdoctoral position in cognitive control and language production Message-ID: <200406030941.58551.oreilly@psych.colorado.edu> Postdoctoral position in cognitive control and language production: Researchers collaborating through an NIMH-funded Interdisciplinary Behavioral Science Center are investigating the role of prefrontal cortex and mechanisms of cognitive control in language production. The research group includes Randy O'Reilly (University of Colorado, Boulder), Jonathan Cohen (Princeton University), Maryellen MacDonald (University of Wisconsin), and Todd Braver (Washington University). The postdoctoral position will be based at either Colorado or Princeton but will involve interaction with investigators at all four sites. The postdoctoral researcher will collaborate on computational and empirical studies of serial ordering and maintaining activation of information during utterance planning. Applicants should have a strong background in language production and computational modeling. Salary will be based on NIH postdoctoral scale, and the position may be renewable beyond the first year. The Ph.D. must be completed before beginning the position. Applicants should send a CV, a statement of research interests, representative preprints/reprints, and arrange for three letters of reference to be sent via email to Randall O'Reilly: oreilly at psych.colorado.edu (PDF format preferred). Applications will be reviewed on an ongoing basis, and will continue until the position is filled. The start date is flexible, but preferably within the coming academic year. For more information and contact addresses, see http://psych.colorado.edu/~oreilly/postdoc_ibsc.html From michael.spratling at kcl.ac.uk Thu Jun 3 13:44:35 2004 From: michael.spratling at kcl.ac.uk (Michael Spratling) Date: Thu, 03 Jun 2004 18:44:35 +0100 Subject: postdoc position, King's College London Message-ID: <40BF6383.8060206@kcl.ac.uk> An enthusiastic and well-qualified post-doctoral researcher is required to develop a biologically inspired neural network model that will be used to explore the effects of inter-regional cortical interactions in visual information processing. The project aims to understand the neural mechanisms underlying cortical region interactions and particularly the effects of top-down information on visual perception. It is hoped that this research will advance understanding of a number of perceptual processes (in areas such as attention, image segmentation, perceptual learning, and categorisation), and in so doing, identify computational principles appropriate for building improved machine vision systems. This project is part of an EPSRC-funded collaboration between the Division of Engineering at King's College London and the Centre for Brain and Cognitive Development (CBCD), Birkbeck College, London. The post-holder will be based in King's College, but will be expected to collaborate closely with researchers in the CBCD, where the predictions of the model will be empirically tested. Applicants should have a proven ability to carry out high quality research, have a genuine interest in the neural mechanisms underlying visual perception, and be keen to develop simulations of both psychological and neuro-physiological data. The successful applicant is expected to have a PhD in a relevant area, have a good knowledge of neural networks and/or computational neuroscience, be comfortable programming in C or C++ in a UNIX environment and ideally should be familiar with MATLAB. Informal inquiries can be directed to Dr Michael Spratling via e-mail: michael.spratling at kcl.ac.uk. The position is available from the 1st October 2004 for a period of 33 months. The starting salary is at SP6 on the RA1A scale, currently =A323,144 per annum inclusive of London Allowance. Further particulars are available by contacting the Personnel Office, King's College London, Strand, London WC2R 2LS, strand-recruitment at kcl.ac.uk, fax 020 7848 1352. Please quote reference W1/CEM/46/04 on all correspondence. The closing date for receipt of completed application forms is: 25 June 2004. Equality of opportunity is College policy. From seth at nsi.edu Mon Jun 7 19:00:41 2004 From: seth at nsi.edu (Anil K Seth) Date: Mon, 07 Jun 2004 16:00:41 -0700 Subject: Workshop on Neurorobotics and Neuroinformatics Message-ID: <5.1.0.14.0.20040607155951.049d4c20@mail.nsi.edu> We are pleased to announce the following workshop, to be held on July 17, 2004 in Santa Monica, CA, as part of the 8th International Conference on the Simulation of Adaptive Behavior (SAB2004): Neurorobotic Models in Neuroscience and Neuroinformatics Adaptive behavior in biological organisms results from interactions among brains, bodies, and environments. We propose in this workshop to explore neurorobotic approaches to understanding these interactions. A key feature of these approaches is the incorporation of aspects of neuroanatomy or neurophysiology that allow comparison with empirical data. More information is available at www.nsi.edu/users/seth/SABwsp.htm We have a great lineup of invited speakers, including: Hiroshi Kimura, Stefan Schaal, Auke Ijspeert, Angelo Arleo, Philippe Gaussier, Jeff Krichmar, Kenji Doya, Ricardo Chavarriaga, Will Alexander, Olaf Sporns, and Steve Potter. We also have a great lineup of coffee breaks, allowing plenty of time for informal discussion. The day will end with dinner somewhere in Santa Monica. We expect this workshop to have a significant impact in both Neuroinformatics (there will be a special issue of the Neuroinformatics journal based on the workshop contents) and in Adaptive Behavior. Come and be a part of it! You can register for only $90 (dinner not included) via the main SAB website: www.isab.org.uk/sab2004/ Please feel free to contact me (seth at nsi.edu) with any questions. Best, Anil Seth Jeff Krichmar Olaf Sporns Auke Ijspeert Ricardo Chavarriaga ----------------------------------------------------------------------------- Anil K Seth, D.Phil., The Neurosciences Institute, 10640 John Jay Hopkins Drive, San Diego, CA 92121, tel: (858) 626 2066, fax: (858) 626 2099 email: seth at nsi.edu, web: www.nsi.edu/users/seth/ From schunn+ at pitt.edu Tue Jun 8 13:45:05 2004 From: schunn+ at pitt.edu (Christian Schunn) Date: Tue, 08 Jun 2004 13:45:05 -0400 Subject: ICCM2004 registration and poster submission info Message-ID: <93AEBAF2-B973-11D8-B937-000393BCD4AC@pitt.edu> International Conference on Cognitive Modeling Pittsburgh, USA Carnegie Mellon University + University of Pittsburgh July 29 - Aug 1, 2004 Online registration for ICCM2004 is now open. Early registration rates are valid until June 28, 2004. Follow the registration link at: http://simon.lrdc.pitt.edu/~iccm/ Early registration: Students $50, Nonstudents $100 Full registration (after June 28th): Students $75, Nonstudents $150. These fees include the cost of the program, the proceedings, the poster reception, the banquet, the break food, and a few other surprises (i.e., the conference fee is surprisingly low). Full and half-day tutorials (on CHREST, Cogent, EPIC, and SOAR) are taking place on the 29th of July, at an extra cost. ($40/half day for nonstudents, $30/half day for students). Registration for tutorials, parking, and staying on campus at the dorms are all done on the same registration page. Those choose to stay at local hotels will make their hotel reservations directly with the hotels. Links are available at: http://simon.lrdc.pitt.edu/~iccm/hotels.html Plenary speakers are Ken Forbus and Michael Mozer. Special symposia include modeling of eye-movements and the pokerbot competition results. The schedule of talks will appear later this week. The submission deadline for 2-page poster-abstracts for ICCM2004 is June 15, 23:59 EDST To submit a poster abstract, go to: http://simon.lrdc.pitt.edu/~iccm/abstract-submit.html Formatting instructions can be found at: http://simon.lrdc.pitt.edu/~iccm/ICCM2004-Submissions.html The schedule of talks will be posted later this week. For further information about ICCM2004, see http://simon.lrdc.pitt.edu/~iccm/ CONFERENCE CHAIRS Marsha Lovett (lovett at cmu.edu) Christian Schunn (schunn at pitt.edu) Christian Lebiere (clebiere at maad.com) Paul Munro (pmunro at mail.sis.pitt.edu) From wolpert at ion.ucl.ac.uk Mon Jun 14 04:44:49 2004 From: wolpert at ion.ucl.ac.uk (Daniel Wolpert) Date: Mon, 14 Jun 2004 09:44:49 +0100 Subject: Postdoctoral Position in Computational Sensorimotor Control Message-ID: A postdoctoral position in human sensorimotor control is available in the laboratory of Daniel Wolpert at the Institute of Neurology, University College London. Ongoing work in the laboratory examines Bayesian and optimal control processes in human motor learning, sensorimotor integration and control. The successful applicant will be expected to conduct independent research involving computationally motivated experimental studies in humans. Applicants should have a PhD and technical expertise and skills relevant to the study of human movement; however, applicants with a strong background in computational neuroscience who wish to learn experimental approaches will also be considered. Experience with Matlab & C++ would be very advantageous. University College London has a large concentration of researchers in the areas of motor control and computational neuroscience, including the neighbouring Gatsby Computational Neuroscience Unit and Institute of Cognitive Neuroscience. Further details of the post and laboratory facilities can be found on www.hera.ucl.ac.uk. Informal enquiries can be addressed by email to Professor Daniel Wolpert (wolpert at ion.ucl.ac.uk). The position is available from September 2004 for one year in the first instance. Starting salary up to =A333,849 pa inclusive, depending on experience, superannuable. Applicants should provide by 30 June 2004 (preferably by email to e.bertram at ion.ucl.ac.uk): - a 1 page statement of research interests - copy of CV (2 if sent by post) - names and contact details of 3 referees - 1 copy of Declaration (required - see further details of posts) - Equal Opportunities form (optional - see further details of posts) to: Miss E Bertram, Assistant Secretary (Personnel) Institute of Neurology Queen Square London WC1N 3BG Fax: +44 (0)20 7278 5069 Email: e.bertram at ion.ucl.ac.uk UCL Taking Action for Equality From l.wiskott at biologie.hu-berlin.de Mon Jun 14 13:15:35 2004 From: l.wiskott at biologie.hu-berlin.de (Laurenz Wiskott) Date: Mon, 14 Jun 2004 19:15:35 +0200 Subject: Postdoc position available in Berlin, Germany Message-ID: <16589.56631.558957.402173@huxley.biologie.hu-berlin.de> A postdoc position will be available January 2005 in the group of Laurenz Wiskott Institute for Theoretical Biology Humboldt-University Berlin for modeling hippocampal function including adult neurogenesis more info at http://itb.biologie.hu-berlin.de/~wiskott/jobs2.html From arno at salk.edu Mon Jun 14 13:37:58 2004 From: arno at salk.edu (Arnaud Delorme) Date: Mon, 14 Jun 2004 10:37:58 -0700 Subject: First EEGLAB Workshop Message-ID: <40CDE276.6090003@salk.edu> Call for pre-registration for the ... First EEGLAB Workshop EEGLAB is a growing open source Matlab environment for signal processing and visualization of EEG, MEG and other electrophysiological signals. For complete information, visit http://sccn.ucsd.edu/eeglab The first three-day EEGLAB workshop will be held Oct. 28-30, 2004, following the Society for Neuroscience meeting in San Diego, at the San Diego Supercomputer Center of the University of California San Diego in La Jolla. Its goals are to introduce EEGLAB to those considering its use in their research and to help current users perform more advanced analyses. Pre-registration applications are being accepted now through July 20. For detailed information and application, visit http://sccn.ucsd.edu/eeglab/workshop/ The workshop faculty, Arnaud Delorme Julie Onton Tzyy-Ping Jung Scott Makeig -- *Arnaud Delorme, Ph.D.* Computational Neurobiology Lab, Salk Institute 10010 North Torrey Pines Road La Jolla, CA 92037 USA *Tel* : /(+1)-858-458-1927 ext 15/ *Fax* : /(+1)-858-458-1847/ *Web page *: www.sccn.ucsd.edu/~arno *To think upon*: Life is either a daring adventure or nothing. /Helen Keller/ From bassis at dsi.unimi.it Tue Jun 15 11:04:33 2004 From: bassis at dsi.unimi.it (Simone Bassis) Date: Tue, 15 Jun 2004 17:04:33 +0200 Subject: WIRN 04 DEADLINE EXTENSION Message-ID: <200406151504.31592.bassis@dsi.unimi.it> [we apologize for multiple receipts] Due to numerous solicitations the deadline for submitting papers to WIRN04 (XV Italian Workshop on Neural Networks, http://siren.dsi.unimi.it/conferences/SI/html) has been delayed to June 26. The electronic submission is regularly open (a temporary inconvenience has been removed). Waiting for you in Perugia, best regards Bruno Apolloni From dgw at MIT.EDU Tue Jun 15 14:27:11 2004 From: dgw at MIT.EDU (David Weininger) Date: Tue, 15 Jun 2004 14:27:11 -0400 Subject: book announcement--Forbes Message-ID: <200406151427116322@outgoing.mit.edu> I thought readers of the Connectionists List might be interested in this book. For more information, please visit http://mitpress.mit.edu/0262062410/ Thank you! Best, David Imitation of Life How Biology Is Inspiring Computing Nancy Forbes As computers and the tasks they perform become increasingly complex, researchers are looking to nature--as model and as metaphor--for inspiration. The organization and behavior of biological organisms present scientists with an invitation to reinvent computing for the complex tasks of the future. In Imitation of Life Nancy Forbes surveys the emerging field of biologically inspired computing, looking at some of the most impressive and influential examples of this fertile synergy. Forbes points out that the influence of biology on computing goes back to the early days of computer science--John von Neumann, the architect of the first digital computer, used the human brain as the model for his design. Inspired by von Neumann and other early visionaries, as well as by her work on the "Ultrascale Computing" project at the Defense Advanced Research Projects Agency (DARPA), Forbes describes the exciting potential of these revolutionary new technologies. She identifies three strains of biologically inspired computing: the use of biology as a metaphor or inspiration for the development of algorithms; the construction of information processing systems that use biological materials or are modeled on biological processes, or both; and the effort to understand how biological organisms "compute," or process information. Forbes then shows us how current researchers are using these approaches. In successive chapters, she looks at artificial neural networks; evolutionary and genetic algorithms, which search for the "fittest" among a generation of solutions; cellular automata; artificial life--not just a simulation, but "alive" in the internal ecosystem of the computer; DNA computation, which uses the encoding capability of DNA to devise algorithms; self-assembly and its potential use in nanotechnology; amorphous computing, modeled on the kind of cooperation seen in a colony of cells or a swarm of bees; computer immune systems; bio-hardware and how bioelectronics compares to silicon; and the "computational" properties of cells. Nancy Forbes works as a science and technology analyst for the federal government. She has advanced degrees both in physics and the humanities, and has served as a contributing editor for The Industrial Physicist and Computing in Science and Engineering. 6 x 9, 176 pp., 48 illus., cloth, ISBN 0-262-06241-0, $25.95 ______________________ David Weininger Associate Publicist The MIT Press 5 Cambridge Center, 4th Floor Cambridge, MA 02142 617 253 2079 617 253 1709 fax http://mitpress.mit.edu From l.s.smith at cs.stir.ac.uk Wed Jun 16 04:45:17 2004 From: l.s.smith at cs.stir.ac.uk (Professor Leslie Smith) Date: Wed, 16 Jun 2004 09:45:17 +0100 Subject: BICS2004, Stirling, Scotland: Call for Participation Message-ID: <40D0089D.3080203@cs.stir.ac.uk> Brain Inspired Cognitive Systems - BICS2004: call for participation. University of Stirling, Scotland, UK August 29 - September 1, 2004 First International ICSC Symposium on Cognitive Neuro Science Chair: Prof. Igor Aleksander, Imperial College London, U.K Second International ICSC Symposium on Biologically Inspired Systems Chair: Prof. Leslie Smith, University of Stirling, U.K. Third International ICSC Symposium on Neural Computation Chair: Dr. Amir Hussain, University of Stirling, U.K. Plenary Speakers: Prof. Rodney Douglas Prof. Graham Hesketh Professor Erkki Oja Owen Holland Prof JG Taylor Prof David Willshaw BICS2004 is being held in Stirling, a historic city located in the heart of Scotland. The conference venue and accommodation are all on the University campus, which is convenient for visiting the ancient castle and Wallace monument, or for going further afield and exploring the beautiful countryside. Register now for this conference: see the website for the full programme and for registration: http://www.icsc-naiso.org/conferences/bics2004/bics-cfp.html -- Professor Leslie S. Smith, Dept of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA, Scotland l.s.smith at cs.stir.ac.uk Tel (44) 1786 467435 Fax (44) 1786 464551 www http://www.cs.stir.ac.uk/~lss/ UKRI IEEE NNS Chapter Chair: http://www.cs.stir.ac.uk/ieee-nns-ukri/ From hinton at cs.toronto.edu Wed Jun 16 10:17:40 2004 From: hinton at cs.toronto.edu (Geoffrey Hinton) Date: Wed, 16 Jun 2004 10:17:40 -0400 Subject: new book on semantic cognition Message-ID: <04Jun16.101742edt.453135-627@jane.cs.toronto.edu> There is a nice new book showing that distributed representations in connectionist networks can explain a lot of psychological data: ___________________________________________ Semantic Cognition A Parallel Distributed Processing Approach Timothy T. Rogers and James L. McClelland http://mitpress.mit.edu/0262182394/ This groundbreaking monograph offers a mechanistic theory of the representation and use of semantic knowledge, integrating the strengths and overcoming many of the weaknesses of hierarchical, categorization-based approaches, similarity-based approaches, and the approach often called "theory theory." Building on earlier models by Geoffrey Hinton in the 1980s and David Rumelhart in the early 1990s, the authors propose that performance in semantic tasks arises through the propagation of graded signals in a system of interconnected processing units. The representations used in performing these tasks are patterns of activation across units, governed by weighted connections among them. Semantic knowledge is acquired through the gradual adjustment of the strengths of these connections in the course of day-to-day experience. The authors show how a simple computational model proposed by Rumelhart exhibits a progressive differentiation of conceptual knowledge, paralleling aspects of cognitive development seen in the work of Frank Keil and Jean Mandler. The authors extend the model to address aspects of conceptual knowledge acquisition in infancy, disintegration of conceptual knowledge in dementia, "basic-level" effects and their interaction with expertise, and many findings introduced to support the idea that semantic cognition is guided by naive, domain-specific theories. From nello at wald.ucdavis.edu Wed Jun 16 19:22:27 2004 From: nello at wald.ucdavis.edu (Nello Cristianini) Date: Wed, 16 Jun 2004 16:22:27 -0700 (PDT) Subject: Book: Kernel Methods for Pattern Analyis Message-ID: <20040616162002.I6178-100000@anson.ucdavis.edu> New book, available now: Kernel Methods for Pattern Analysis John Shawe-Taylor & Nello Cristianini Cambridge University Press, 2004 ISBN: 0521813972 http://www.kernel-methods.net/ Kernel methods provide an approach to pattern analysis that has developed over the last 12 years since the introduction of Support Vector Machines. This new book has two main aims: to provide an introduction to this approach and act as a cook-book for practitioners. The three parts, Concepts, Algorithms and Kernels, contain dozens of algorithms and kernels, pseudocode and detailed derivations, while Matlab code is freely available from the book website. The book provides a comprehensive account of the field of Kernel Methods for Pattern Analysis, summarizing more than a decade of expansion of these exciting and successful techniques. From nipsinfo at salk.edu Thu Jun 17 17:22:30 2004 From: nipsinfo at salk.edu (nipsinfo@salk.edu) Date: Thu, 17 Jun 2004 14:22:30 -0700 Subject: NIPS 2004 Survey Message-ID: To: NIPS Attendees and Connectionists: We are asking for a few minutes of your time to complete an online survey consisting of seven questions. The survey deals with a some major changes to the format of the 2004 Neural Information Processing Systems Conference currently under consideration. The impetus for these possible changes is to accommodate the growth in submissions in recent years, and the diverse demographics of Conference attendees. One concern that has been raised regarding the increased size of the meeting is the unwieldy scale of the Poster Sessions. Another is that while the representation in machine learning at NIPS is growing, many feel that its focus is becoming narrower; the result of this has been that many prominent computational neuroscientists, cognitive modelers, etc. have stopped attending NIPS because they believe the meeting to be overly dominated by machine learning. The survey can be found at the NIPS 2004 website: https://register.nips.salk.edu/surveys/survey.php?id=3 Your responses to these questions, as well as any additional input you may have, will be very valuable in shaping the future of NIPS, and we very much appreciate your participation. Neural Information Processing Systems Conference 2004 From bogus@does.not.exist.com Thu Jun 17 11:55:02 2004 From: bogus@does.not.exist.com () Date: Thu, 17 Jun 2004 16:55:02 +0100 Subject: 9th Neural Computation and Psychology Workshop: Extended deadline and Limited Travel grants Message-ID: <736F0925D69F9941B3BA8AEED0F5E75C029A746B@02-CSEXCH.uopnet.plymouth.ac.uk> From d.mareschal at bbk.ac.uk Thu Jun 17 09:29:19 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Thu, 17 Jun 2004 14:29:19 +0100 Subject: connectionism and development Message-ID: Dear all, The following special section in Developmental Science 7(2) may be of interest to readers of this list. SECIAL SECTION Beyond Backprogation: emerging trends in connectionist models of 131 development Matthew Schlesinger and Domenico Parisi Autoassociator networks: insights into infant cognition 133 Sylvain Sirois Hebbian learning and development 141 Yuko Munakata and Jason Pfaffly Modeling developmental transitions in adaptive resonance theory 149 Maartje E. J. Raijmakers and Peter C. M. Molenaar Evolving agents as a metaphor for the developing child 158 Matthew Schlesinger Commentary 165 David Klahr This issue also contains a scientific obituary for Elizabeth Bates ================================================= Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================= From arthur at tuebingen.mpg.de Fri Jun 18 21:43:55 2004 From: arthur at tuebingen.mpg.de (Arthur Gretton) Date: 19 Jun 2004 03:43:55 +0200 Subject: extended deadline for ICML/COLT session on kernel methods Message-ID: <1087609435.1030.5.camel@bertie> A non-text attachment was scrubbed... Name: not available Type: multipart/alternative Size: 919 bytes Desc: not available Url : https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/0200062004/17d9c47f/attachment-0001.bin From special at math.unipd.it Fri Jun 18 05:17:04 2004 From: special at math.unipd.it (Alessandro Sperduti) Date: Fri, 18 Jun 2004 11:17:04 +0200 (CEST) Subject: Cfp: Special Issue Message-ID: ---------------------------------------------------------------------- C A L L F O R P A P E R S Special Issue of Neural Networks on Neural Networks and Kernel Methods for Structured Domains Co-editors:Barbara Hammer, Craig Saunders, Alessandro Sperduti (http://www.informatik.uni-osnabrueck.de/barbara/specialNN/) ---------------------------------------------------------------------- In recent years several researchers have started to consider the adaptive processing of structured data. This interest is motivated by two main reasons: i) several very important computational problems in Bioinformatics, Chemistry, document classification and filtering (just to name a few), require the use of some Machine Learning procedure to be properly treated because their complexity does not allow a formal and precise definition of the problem and thus no algorithmic solution to the problem is known; on the other hand experimental data is available and usable by a Machine Learning approach; ii) in many of the above problems, the objects of interest are more naturally represented via varying-size structured representations, such as sequences, strings, trees, directed or undirected graphs. Moreover, it is commonly believed that any computational process representing these objects via flat fixed-size vectorial representations, risks to discard structural information relevant for solving the task at hand. The aim of the special issue is to bring together recent works developed mainly into the fields of Recurrent and Recursive Neural Networks, and Kernel Methods, with the hope to function as a conceptual support for the improvement of current technologies and the growth of new integrated methodologies for learning in structured domains. TOPICS: Recurrent and Recursive Neural Networks, Self-Organizing Maps for Structures, Support Vector Machines and Kernel Methods for Structured Domains, Fisher Kernels, Probabilistic Models for Structured Domains, Applications of the above methods to Structured Domains, other relevant topics. ---------------------------------------------------------------------- I M P O R T A N T D A T E S Deadline for submission October 11, 2004 Notification of acceptance: March 14, 2005 Deadline for final paper: June 6, 2005 ---------------------------------------------------------------------- P A P E R S U B M I S S I O N Format: http://www.informatik.uni-osnabrueck.de/barbara/specialNN/ Papers must be submitted in electronic form (either Postscript or PDF) to the following e-mail address: specialNN.structure at math.unipd.it Hard-paper submissions are accepted only if the authors do not have access to electronic facilities, and should be mailed to the following address: Prof. Alessandro Sperduti Dipartimento di Matematica Pura ed Applicata Via Belzoni 7 35131 Padova ITALY ---------------------------------------------------------------------- From scheler at stanford.edu Sat Jun 19 22:26:15 2004 From: scheler at stanford.edu (Gabriele Scheler) Date: Sat, 19 Jun 2004 19:26:15 -0700 Subject: Papers on Neural Plasticity Message-ID: <1087698375.40d4f5c75704a@webmail.stanford.edu> Dear colleagues, Here is an announcement for two recent papers (and an earlier one from 2003) on non-LTP-like brain plasticity. available from arxiv.org Summary: ------------------------------ The evidence for learning and memory expressed by the distribution of receptors for other neurochemicals, prominently dopamine, acetylcholine etc. is reviewed. Theoretical models are outlined (a) for synaptic gating by presynaptic receptor activity and (b) for intrinsic, whole-neuron adaptivity in membrane excitation ('transfer') functions. The upshot is that learning and memory may very likely be expressed by the adjustment of G-protein coupled receptors, even though the amount of heterogeneity (storage capacity) and the amount of long-term persistence of adaptivity are not well investigated yet. Known to be the target of many psychoactive substances, and prominent in addiction, fear and social learning, we conjecture that neuromodulatory plasticity has a quite general function in 'emotional learning' in many parts of the brain. --------------------------------- Regulation of Neuromodulator Efficacy - Implications for Whole-Neuron and Synaptic Plasticity. Progress in Neurobiology 72(6), April 2004. http://www.arxiv.org/pdf/q-bio.NC/0403011 Memorization in a neural network with adjustable transfer function and conditional gating. online since March 2004. http://www.arxiv.org/pdf/q-bio.MN/0401039 Presynaptic modulation as fast synaptic switching: state-dependent modulation of task performance. Proceedings of IJCNN'03, Portland, Oregon, June 2003. http://www.arxiv.org/pdf/cs.NE/0401020 -------------------------------- Dr. G. Scheler Research Scientist ISLE/CSLI 25 Ventura Hall Stanford, Ca. 94305 http://www.stanford.edu/~scheler From auke.ijspeert at epfl.ch Tue Jun 22 03:57:11 2004 From: auke.ijspeert at epfl.ch (Auke Jan Ijspeert) Date: Tue, 22 Jun 2004 09:57:11 +0200 Subject: SAB2004: calls for participation and abstracts (ext'd deadline) Message-ID: <40D7E657.3070601@epfl.ch> Dear Connectionists, Researchers in adaptive behavior, neural computation, and artificial neural networks applied to robotics might be interested in the next SAB2004, From Animals to Animats 8 conference that will take place in Santa Monica July 13-17 2004. See the call for participation and the call for abstracts below. Note that the deadline for the Last-Minute-Results abstract submission has been extended until July 9th. Best regards, Auke Ijspeert -----SAB2004 Call for participation and abstracts (ext'd deadline)----- SAB2004, From Animals to Animats 8, The Eighth International Conference on the Simulation of Adaptive Behavior Santa Monica (Los Angeles), July 13-17 2004 http://www.isab.org/sab04/ ----- Call for participation ----- The objective of this interdisciplinary conference is to bring together researchers in computer science, artificial intelligence, alife, control, robotics, neurosciences, ethology, and related fields so as to further our understanding of the behaviors and underlying mechanisms that allow natural and artificial animals to adapt and survive in uncertain environments. The conference will focus on experiments with well-defined models --- robot models, computer simulation models, mathematical models --- designed to help characterize and compare various organizational principles or architectures underlying adaptive behavior in real animals and in synthetic agents, the animats. The preliminary technical program can be found at http://www.isab.org/sab04/program/ See http://www.isab.org/sab04/ for registration and more details. -----Call for abstracts for the Last-Minute-Results poster session----- SAB2004 introduces a new feature: the Last-Minute-Results poster session (July 9th, extended deadline). This special session will offer researchers in Adaptive Behavior the opportunity to present their most recent results at SAB2004 (http://www.isab.org/sab04/). The goal of this session is to provide an informal setting in which participants can reveal and discuss their latest results and developments at the time of the conference (13-17 July 2004). Researchers are invited to submit a two-page abstract describing recent results/developments. Abstracts should be sent electronically as a PDF file to the conference general email address, sab2004 at isab.org, with "Last Minute Results" in the email Subject line. The call for abstracts is open to all (i.e. to people both with and without an accepted paper at the conference). There is a limit of one abstract/poster per participant as a first author. After review by the conference chairs, authors of accepted abstracts will be allowed to present their results as a poster in a special Last-Minute-Results session. Note that the abstracts will not be considered as publications and will not be included in the conference proceedings. This also means that you keep the abstracts' copyrights. Deadlines: July 9th: Abstract submission deadline (but note that there is a limited number of slots, and that posters will be accepted after a short review on a first come, first served basis) July 13-17 2004: SAB'04 conference Contributions treating any of the following topics from the perspective of adaptive behavior will receive special emphasis: The Animat approach Characterization of agents and environments Passive and active perception Motor control Visually-guided behaviors Action selection Behavioral sequencing Navigation and mapping Internal models and representation Learning and development Motivation and emotion Collective and social behavior Emergent structures and behaviors Neural correlates of behavior Evolutionary and co-evolutionary approaches Autonomous robotics Humanoid robotics Software agents and virtual creatures Applied adaptive behavior Animats in education Philosophical and psychological issues X-Antivirus-NBU.BG-Mail-From: office at cogs.nbu.bg via anna-karenina X-Antivirus-NBU.BG: 1.22st (Clear:RC:1(192.168.50.47):. Processed in 0.028888 secs Process 29365) From office at cogs.nbu.bg Wed Jun 23 13:26:59 2004 From: office at cogs.nbu.bg (Cogs Office) Date: Wed, 23 Jun 2004 10:26:59 -0700 Subject: Cognitive Science, NBU - Announcement Message-ID: THE CENTRAL AND EAST EUROPEAN CENTER FOR COGNITIVE SCIENCE INVITES APPLICATIONS FOR: 1. One-year post-doctoral fellowships for the 2003/2004 academic year (with a possible extension); 2. Two-year Ph.D. studentships (starting in October 2004). For more information: http://www.nbu.bg/cogs/events/2004/positions.html Central and East European Center for Cognitive Science New Bulgarian University 21 Montevideo Str. Sofia 1618, Bulgaria phone: (+3592) 8110-401 e-mail: office at cogs.nbu.bg From nips04pub at hotmail.com Thu Jun 24 16:25:04 2004 From: nips04pub at hotmail.com (John Platt) Date: Thu, 24 Jun 2004 13:25:04 -0700 Subject: NIPS 2004 Call for Demonstrations Message-ID: CALL FOR DEMONSTRATIONS Neural Information Processing Systems --- Natural and Synthetic Monday, December 13 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc DEADLINE FOR DEMONSTRATION PROPOSALS: August 1, 2004 Based on their success in 2002 and 2003, the Neural Information Processing Systems Conference will again include a separate track for Demonstrations. The Demonstrations will take place in parallel with the NIPS poster sessions on Tuesday and Wednesday evenings (December 14 and 15). The key requirements for Demonstrations is that they be LIVE and INTERACTIVE and that they present a compelling view of an emerging technology. Past Demonstrations have covered a very wide range. Areas of interest for the Demonstrations track have previously included the following: analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, robotics, bioMEMS (microelectromechanical systems), biomedical instrumentation, neural prostheses, photonics, real-time multimedia systems, large-scale neural emulators, software Demonstrations of novel algorithms, and open-source software toolboxes. NIPS is a continually-evolving interdisciplinary Conference, which attracts cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The Demonstration track enables researchers to highlight scientific advances, systems, and technologies in ways that go beyond conventional poster presentations. It will provide a unique forum for demonstrating advanced technologies (hardware and software), and fostering the direct exchange of knowledge. We hope that this track will stimulate interactions between researchers from different fields or approaches. Submissions accepted in the Demonstrations track will be published on the NIPS web site, but will not appear in printed proceedings. However, submitting your work to the Demonstration track does not preclude the submission of a companion paper to the regular NIPS Conference; joint submissions are very much encouraged. We also encourage authors submitting Demonstrations to consider organizing a Workshop at NIPS 2004. There will be a separate room for these Demonstrations and participants will have access to power strips, tables and poster boards. Monitors will also be provided on request at their rental cost. Participants are responsible for ensuring that their Demonstration is sufficiently portable; additional hardware beyond that specified above might be provided at cost, if readily available. SUBMISSION INSTRUCTIONS: Proposals for Demonstrations will be reviewed by the Demonstrations Co-Chairs. Demonstration proposals should be submitted via the web form available at http://nips.cc. Demonstrators will be asked to enter information about the nature of the Demonstration, in particular they will be asked to describe first the user experience and then the underlying technology. Proposals that are simply papers in disguise will be rejected, this session is for live, interactive experiences that compellingly demonstrate new technology. The Demonstration track is not an alternative poster session. Proposers will also be asked about the present state of their Demonstration in order that the co-chairs may judge whether the Demonstration can actually be made functional. Past experience has shown that simpler Demonstrations that make one point are usually more interesting to attendees. Complex Demonstrations involving multiple technologies and partners have not been as effective. NIPS 2004 DEMONSTRATIONS CO-CHAIRS: Tobi Delbruck, Institute of Neuroinformatics, University of Zurich, ETH Zurich; Timmer Horiuchi, Electrical and Computer Engineering Department, Institute for Systems Research and Neuroscience and Cognitive Science Program, University of Maryland. DEADLINE FOR DEMONSTRATION PROPOSALS: August 1, 2004 From mjhealy at chango.eece.unm.edu Fri Jun 25 21:02:48 2004 From: mjhealy at chango.eece.unm.edu (M. Healy) Date: Fri, 25 Jun 2004 19:02:48 -0600 (MDT) Subject: Technical report available Message-ID: A technical report on a mathematical model of the semantics of neural networks is now available on the Dspace archive in the collection 2004. To access it, go to the University of New Mexico site at https://repository.ece.unm.edu/dspace/ and search on http://hdl.handle.net/1928/33 . Here's an abstract: Neural Networks, Knowledge, and Cognition: A Mathematical Semantic Model Based upon Category Theory, by M. J. Healy and T. P. Caudell Category theory can be applied to mathematically model the semantics of cognitive neural systems. We discuss semantics as a hierarchy of concepts, or symbolic descriptions of items sensed and represented in the connection weights distributed throughout a neural network. The hierarchy expresses subconcept relationships, and in a neural network it becomes represented incrementally through a Hebbian-like learning process. The categorical semantic model described here explains the learning process as the derivation of colimits and limits in a concept category. It explains the representation of the concept hierarchy in a neural network at each stage of learning as a system of functors and natural transformations, expressing knowledge coherence across the regions of a multi-regional network equipped with multiple sensors. The model yields design principles that constrain neural network designs capable of the most important aspects of cognitive behavior. Please let me know right away if there are any problems. I'll warmly receive any questions or comments. - Mike From terry at salk.edu Sat Jun 26 12:32:25 2004 From: terry at salk.edu (Terry Sejnowski) Date: Sat, 26 Jun 2004 09:32:25 -0700 (PDT) Subject: NEURAL COMPUTATION 16:8 In-Reply-To: <200405242248.i4OMmXL28618@dax.snl.salk.edu> Message-ID: <200406261632.i5QGWPO06371@dax.salk.edu> Neural Computation - Contents - Volume 16, Number 8 - August 1, 2004 LETTERS A Coarse-to-Fine Disparity Energy Model with Both Phase-Shift and Position-Shift Receptive Field Mechanisms Yuzhi Chen and Ning Qian A Preference for Phase-Based Disparity in a Neuromorphic Implementation of the Binocular Energy Model Eric K.C. Tsang and Bertram E. Shi Learning Classification in the Olfactory System of Insects Ramon Huerta, Thomas Nowotny, Marta Garcia-Sanchez, H.D.I Abarbanel, and M.I. Rabinovich Adaptive Blind Separation with Unknown Number of Sources Ji-Min Ye, Xiao-Long Zhu, and Xian-Da Zhang Unsupervised Spike Detection and Sorting with Wavelets and Superparamagnetic Clustering R. Quian Quiroga, Z. Nadasdy and Y. Ben-Shaul Decomposition Methods for Linear Support Vector Machines Wei-Chun Kao, Kai-Min Chung, Chia-Liang Sun, and Chih-Jen Lin An Asymptotic Statistical Theory of Polynomial Kernel Methods Kazushi Ikeda A Neural Root Finder of Polynomials Based on Root Moments De-Shuang Huang, Horace H. S. Ip, and Zheru Chi ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2004 - VOLUME 16 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $108 $54 $57.78 Individual $95 $101.65 $143 $85 $90.95 Institution $635 $679.45 $689 $572 $612.04 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From pmr at di.fc.ul.pt Tue Jun 29 04:21:37 2004 From: pmr at di.fc.ul.pt (pmr@di.fc.ul.pt) Date: Tue, 29 Jun 2004 09:21:37 +0100 Subject: Extension to Delta Rule already studied before? Message-ID: <1088497297.40e1269148dc5@webservices.di.fc.ul.pt> Dear connectionists, while working on my Ph.D. thesis, I've been using neural nets that apply the 'Delta Rule'. Nevertheless, for some instances of the problems I've been addressing, the 'Delta Rule' does not seem appropriate: it promotes weights that do not differ substantially. While trying to overcome this property, I've came across the following rule: w_ij = w_ij + a_i * (t_j - a_j) * LR * w_ij where... w_ij - weight of the connection from input neuron i to output neuron j a_i - activity of input neuron i t_j - target activity for output unit j a_j - activity of output neuron j LR - learning rate Has this learning rule been studied before? Are there references ? Many thanks for your attention, kind regards, pedro