From nips04pub at hotmail.com Mon May 3 18:56:54 2004 From: nips04pub at hotmail.com (John Platt) Date: Mon, 3 May 2004 15:56:54 -0700 Subject: NIPS 2004 Call for Papers Message-ID: CALL FOR PAPERS --- NIPS 2004 Neural Information Processing Systems --- Natural and Synthetic Monday, December 13 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc Deadline for Paper Submissions: June 4, 2004 Submissions are solicited for the eighteenth annual meeting of an interdisciplinary NIPS Conference (December 14-16) which brings together researchers interested in all aspects of neural and statistical processing and computation. The Conference will include invited talks as well as Oral and Poster Presentations of refereed papers and Demonstrations. It is single track and highly selective. Preceding the main Conference will be one day of Tutorials (December 13), and following it will be two days of Workshops at Whistler/Blackcomb ski resort (December 17-18). INVITED SPEAKERS: * John Donoghue, Brown University --- Mind over Movement: Developing Neurotechnologies to Restore Lost Function; * Gerd Gigerenzer, Max Planck Institute for Human Development --- Fast and Frugal Heuristics: The Adaptive Toolbox; * Nati Linial, Hebrew University of Jerusalem --- Expanders, Eigenvalues and All That; * Bernard Palsson, University of California, San Diego, Systems Biology --- Bringing Genomes to Life; * Shimon Ullman, Weizmann Institute of Science --- Classification, Recognition and Segmentation Using Fragments Hierarchy TUTORIAL SPEAKERS: * William Bialek, Princeton University --- Optimization Principles in Neural Coding and Computation; * Shahar Mendelson, Australian National University --- A Geometric Approach to Statistical Learning Theory; * Radford Neal, University of Toronto --- Bayesian Methods in Machine Learning; * David Parkes, Harvard University --- Computational Mechanism Design and Auctions * Richard Szeliski, Microsoft Research --- Acquiring Detailed 3D Models from Images and Video; * Daniel Wolpert, University College London --- Probabilistic Computations in Human Sensoriomotor Control SUBMISSIONS Papers are solicited in all areas of neural information processing, including (but not limited to) the following: * Algorithms and Architectures --- statistical learning algorithms, neural networks, kernel methods, graphical models, Gaussian processes, independent component analysis, model selection, combinatorial optimization. * Applications --- innovative applications or fielded systems that use machine learning, including systems for time series prediction, bioinformatics, text/web analysis, multimedia processing, and robotics. * Brain Imaging --- neuroimaging, cognitive neuroscience, EEG (electroencephalogram), ERP (event related potentials), MEG (magnetoencephalogram), fMRI (functional magnetic resonance imaging), brain mapping, brain segmentation. * Cognitive Science and Artificial Intelligence --- theoretical, computational, or experimental studies of perception, psychophysics, human or animal learning, memory, reasoning, problem solving, language, and neuropsychology. * Control and Reinforcement Learning --- decision and control, exploration, planning, navigation, Markov decision processes, game-playing, multi-agent coordination, computational models of classical and operant conditioning. * Emerging Technologies --- analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, microrobotics, bioMEMS, neural prostheses, photonics, molecular and quantum computing. * Learning Theory --- generalization and regularization, information theory, statistical physics of learning, Bayesian methods, approximation bounds, online learning and dynamics. * Neuroscience --- theoretical and experimental studies of processing and transmission of information in biological neurons and networks, including spike train generation, synaptic modulation, plasticity and adaptation. * Speech and Signal Processing --- recognition, coding, synthesis, denoising, segmentation, source separation, auditory perception, psychoacoustics, dynamical systems, recurrent networks, Markov models. * Visual Processing --- biological and machine vision, image processing and coding, segmentation, object detection and recognition, motion detection and tracking, visual psychophysics, visual scene analysis and interpretation. * Demonstrations --- Authors wishing to submit to the Demonstration track should consult the Conference web site. REVIEW CRITERIA: Submissions will be refereed on the basis of technical quality, novelty, significance, and clarity. Authors new to NIPS are particularly encouraged to submit. There will be an opportunity after the meeting to revise accepted manuscripts. PAPER FORMAT: Submissions may be up to eight pages in length, including figures and references, using a font no smaller than 10 point. Text is to be confined within a 8.25 inch by 5 inch rectangle. Submissions violating these guidelines will not be considered. SUBMISSION INSTRUCTIONS: NIPS accepts only electronic submissions in postscript and PDF format. The Conference web site will accept electronic submissions from May 19, 2004 until midnight, June 4, 2004, Pacific daylight time. ORGANIZING COMMITTEE: * General Chair, Lawrence Saul, University of Pennsylvania; * Program Chair, Yair Weiss, Hebrew University of Jerusalem; * Tutorials Chair, Richard Zemel, University of Toronto; * Workshops Co-chairs, Satinder Singh Baveja, University of Michigan, Daniel Lee, University of Pennsylvania; * Demonstrations Chairs, Tobi Delbruck, ETH/University of Zurich, Timothy Horiuchi, University of Maryland; * Publications Chair, Leon Bottou, NEC Research Institute; * Publicity Chair, John Platt, Microsoft Research; * Online Proceedings Chair, Andrew McCallum, University of Massachusetts Amherst; * Volunteers Chair, Sam Roweis, University of Toronto. PROGRAM COMMITTEE: Yair Weiss (Chair), Hebrew University of Jerusalem; Yoshua Bengio, University of Montreal; Jeff Bilmes, University of Washington; Nello Cristianini, University of California Davis; Trevor Darrell, Massachusetts Institute of Technology; Geoff Gordon, Carnegie Mellon University; Daphne Koller, Stanford University; John Lafferty, Carnegie Mellon University; Shih-Chii Liu, Swiss Federal Institute of Technology Zurich; Ron Meir, Technion, Israeli Institute of Technology; Maneesh Sahani, University College London; Bernhard Schoelkopf, Max-Planck Institute for Biological Cybernetics; Matthias Seeger, University of California, Berkeley; Richard Shiffrin, Indiana University; Eero Simoncelli, New York University; Martin Wainwright, University of California Berkeley; Daphna Weinshall, Hebrew University of Jerusalem. DEADLINE FOR PAPER SUBMISSIONS: JUNE 4, 2004 From jebara at cs.columbia.edu Tue May 4 13:23:12 2004 From: jebara at cs.columbia.edu (Tony Jebara) Date: Tue, 4 May 2004 13:23:12 -0400 Subject: ICDL 2004 - Extension to May 21, 2004 Message-ID: ICDL 2004 CALL FOR PAPERS DEADLINE HAS BEEN EXTENDED TO FRIDAY, MAY 21 2004 There will be absolutely no further extensions. THIRD INTERNATIONAL CONFERENCE ON DEVELOPMENT AND LEARNING: DEVELOPING SOCIAL BRAINS The Salk Institute October 20-22, 2004 San Diego, California A Satellite Conference preceding The Annual Society for Neuroscience Conference http://www.icdl.cc The goal of the conference is to bring together leading researchers in neuroscience, machine learning, robotics, and developmental psychology, in order to gain new insights about learning and development in natural organisms and robots. The scope of developmental processes to be considered is broad, including cognitive, social, emotional, and many other skills exhibited by humans, and animals. The theme of the conference this year will be "Developing Social Brains", but other topics related to development and learning are welcome. PAPER SUBMISSION The extended submission deadline is May 21, 2004. Papers for the meeting can be submitted ONLY through the conference's web site at: http://www.icdl.cc. Papers can be submitted either as a 200 word summary or as a full paper (max 8 typeset pages). IMPORTANT: There will be NO further submission deadline extensions. SPECIAL ISSUE ON NEUROCOMPUTING Authors of selected papers will be invited to submit an extended version of their paper for publication in a special issue of the Neurocomputing Journal, on Development, Learning, and the Social Brain, published by Elsevier Science B.V. (http://www.elsevier.nl/locate/neucom) INVITED TALKS (Not yet confirmed) John Allman Dana Ballard Rodney Brooks Eric Courchesne Peter Dayan Jeff Elman William Greenough James L. McClelland Pietro Perona Terrence Sejnowski Joan Stiles John Watson REVIEW PROCESS All submitted papers will be reviewed by the program committee. Papers will be judged and accepted for the meeting based on the clarity with which the work is described and the relevance to the goals of the conference. All accepted papers not selected for oral talks as well as papers explicitly submitted as poster presentations will be included in one of three evening poster sessions. Authors will be notified of the presentation format of their papers by the beginning of August. ORGANIZING COMMITTEE: General Chair: Javier R. Movellan: Co-Chairs: Andrea Chiba, Gedeon Deak, Jochen Triesch. Program Chair: Jochen Triesch and Tony Jebara. Program Co-Chairs: Marian Stewart-Bartlett, Gwen Ford Littlewort. Publications Chair: Gedeon Deak. ADVISORY BOARD: Jeff Elman Pat Langley James L. McClelland Sandy Pentland Terrence Sejnowski Mriganka Sur Esther Thelen Juyang Weng PROGRAM COMMITTEE: Minoru Asada Dana Ballard Luis Baumela Simon Baron-Cohen Mark Baxter Jeff Cohn Kerstin Dautenhahn Kenji Doya Martha Farah Teresa Farroni Masahiro Fujita Ann Graybiel William Greenough Michael Hasselmo Shoji Itakura Hiroshi Ishiguro Robert Jacobs David Kleinfeld Mark Konishi Denis Mareshal Risto Miikulainen Douglas Nitz Roz Picard Steven Quartz Rajesh Rao Matthew Schlesinger Gregor Schoener Geoffrey Schoenbaum Linda Smith Olaf Sporns Luc Steels Valerie Stone Manuela Veloso Paul Verschure Christoph von der Malsburg Hiroyuki Yano CONFERENCE REGISTRATION: Register online at http://www.icdl.cc Student Registration is: $150 Non-student Registration is: $290 From cindy at bu.edu Wed May 5 10:13:11 2004 From: cindy at bu.edu (Cynthia Bradford) Date: Wed, 5 May 2004 10:13:11 -0400 Subject: 8th ICCNS: Final Call for Registration Message-ID: <00a801c432ab$18f19b20$903dc580@cnspc31> Apologies if you receive more than one copy of this email. ***** FINAL CALL FOR REGISTRATION ***** EIGHTH INTERNATIONAL CONFERENCE ON COGNITIVE AND NEURAL SYSTEMS May 19-22, 2004 http://cns.bu.edu/cns-meeting/conference.html Boston University CNS Department 677 Beacon Street Boston, Massachusetts 02215 Sponsored by Boston University's Center for Adaptive Systems and Department of Cognitive and Neural Systems with financial support from the Office of Naval Research This single-track conference, which offers invited lectures and contributed lectures and posters, focuses on two main themes: How does the brain control behavior? How can technology emulate biological intelligence? CONFIRMED INVITED SPEAKERS: Tutorial Lecture Series: Stephen Grossberg Keynote Lectures: John Anderson and Miguel Nicolelis Invited Speakers: Ehud Ahissar, Alan D. Baddeley, Moshe Bar, Gail A. Carpenter, Stephen Goldinger, Daniel Kersten, Stephen M. Kosslyn, Tai-Sing Lee, Eve Marder, Bartlett W. Mel, Jeffrey D. Schall, Chantal Stern, Mriganka Sur, Joseph Z. Tsien, William H. Warren Jr., Jeremy Wolfe. Please visit the web site for conference details, including: --the registration form http://cns.bu.edu/cns-meeting/registration.html --a schedule of the oral and poster presentations http://cns.bu.edu/cns-meeting/schedule.html --details about the tutorial lecture series http://cns.bu.edu/cns-meeting/grossberg-lectures.html --local lodging options http://cns.bu.edu/cns-meeting/hotels.html From gregoire.dubois at jrc.it Thu May 6 03:19:38 2004 From: gregoire.dubois at jrc.it (Gregoire Dubois) Date: Thu, 6 May 2004 09:19:38 +0200 Subject: Spatial Interpolation Comparison exercise: who is going to do the "best map"? Message-ID: <000001c4333a$7daf0b80$9badbf8b@ei.jrc.it> Please, find hereafter the announcement for the following international exercise. I would very much appreciate the contribution of the community of connectionists. SIC2004 is the second edition of a scientific exercise dedicated to decision support systems and spatial statistics. SIC2004 stands for Spatial Interpolation Comparison 2004. Participants will receive a subset of an environmental data set (typically measurements of an environmental variable + spatial coordinates of the sampling places) and will have to estimate the values taken by the variable at the remaining locations of the full data set. The true values found at these locations will be made public only at the end of the exercise. Various criteria will be used to assess the performances of the interpolation algorithms (time of calculation, minimum errors, etc.). This edition will focus on automatic mapping algorithms: participants to SIC2004 will have to prepare their algorithms before receiving the data (only sampling locations will be given) and no interaction with the algorithm will be allowed during the exercise. Participants to SIC2004 will be invited to submit a manuscript at the end of the exercise for publication in the online journal GIDA (Geographic Information and Decision Analysis) as well as in a European Report (hardcopy). For more information, please visit the web site www.ai-geostats.org Thank you very much for your attention, Gregoire Keywords: geostatistics,spatial statistics, neural networks, spatial interpolation, GIS. PS: in the case you would like to support such an exercise, could you please forward this information to other potential scientific communities (e.g. machine learning, neural networks, statisticians, etc.). Thank you very much. __________________________________________ Gregoire Dubois (Ph.D.) JRC - European Commission Radioactivity Environmental Monitoring TP 441, Via Fermi 1 21020 Ispra (VA) ITALY Tel. +39 (0)332 78 6360 Fax. +39 (0)332 78 5466 Email: gregoire.dubois at jrc.it WWW: http://www.ai-geostats.org WWW: http://rem.jrc.cec.eu.int From wsenn at cns.unibe.ch Fri May 7 09:06:32 2004 From: wsenn at cns.unibe.ch (Walter Senn) Date: Fri, 07 May 2004 15:06:32 +0200 Subject: Postdoc and PhD position available in computational neuroscience Message-ID: <409B89D8.5010007@cns.unibe.ch> A postdoc and a PhD position is available by October 1, 2004, in the computational neuroscience group at the Institute of Physiology, University of Bern, Switzerland. We are studying biophysical models of neurons and synapses in the context of cortical information processing and learning. The aim of our reaearch is to relate dynamic properties of isolated neurons and synapses (studied in in vitro experiments by the groups of H.-R. Luescher and M.E. Larkum in our Institute) to higher cognitive functions. In our recent work we described slow adaptation of cortical neurons as measured in vitro in our lab [1], and related it to the cortical representation of time [2]. Other experimental and theoretical research in our lab shows that top-down input onto dendrites of layer 5 pyramidal cells can modulate the gain of these neurons [3], as observed in attentional experiments in vivo. In a current project we are studying spike-timing dependent plasticity (STDP) in the context of pattern classification and reinforcement learning [4]. These algorithms will be tested against data from in vitro experiments on long-term plasticity performed in our lab. Ideal candidates have a strong background in neural networks and/or computational neuroscience, and are interested to work at the interface between computational theories and electrophysiology. Please send CV, addresses of three references, and a half page description of your work and interests to wsenn at cns.unibe.ch (but no doc-files, if possible). Walter Senn and Stefano Fusi http://www.cns.unibe.ch/ [1] Neocortical pyramidal cells respond as integrate-and-fire neurons to in vivo-like input currents. A. Rauch, G. La Camera, H-R L?scher, W. Senn and S. Fusi. Journal of Neurophysiology, 90:1598-1612 (2003) http://www.cns.unibe.ch/publications/ftp/paper_transferfunction.pdf [2] Climbing neuronal activity as an event-based cortical representation of time. J. Reutimann, V. Yakovlev, S. Fusi and W. Senn. The Journal of Neuroscience, 24:3295-3303 (2004) http://www.cns.unibe.ch/publications/ftp/paper_climbing_activity.pdf [3] Top-down dendritic input increases the gain of layer 5 pyramidal neurons. M. Larkum, W. Senn and H.-R. L?scher, Cerebral Cortex, in press (2004) http://www.cns.unibe.ch/publications/ftp/paper_gain_modulation04.pdf [4] Learning with bounded synapses generates synaptic democracy and balanced neurons. With a proof of the perceptron convergence theorem for bounded synapses. W. Senn and S. Fusi, submitted (2004) http://www.cns.unibe.ch/publications/ftp/paper_bounded_synapses04.pdf -- Walter Senn Phone office: +41 31 631 87 21 Physiological Institute Phone home: +41 31 332 38 31 University of Bern Fax: +41 31 631 46 11 Buehlplatz 5 Email: wsenn at cns.unibe.ch CH-3012 Bern http://www.cns.unibe.ch/~wsenn/ From dejong at cs.uu.nl Tue May 11 08:06:33 2004 From: dejong at cs.uu.nl (Edwin de Jong) Date: Tue, 11 May 2004 14:06:33 +0200 (MEST) Subject: Preprint available: Ideal Evaluation from Coevolution Message-ID: Dear colleagues, I would like to announce a preprint of the following article: De Jong, E.D. and Pollack, J.B. (2004). Ideal Evaluation from Coevolution. Evolutionary Computation, Vol. 12, no. 2 (to appear). Available from: http://www.cs.uu.nl/~dejong/#coevec Abstract: In many problems of interest, performance can be evaluated using tests, such as examples in concept learning, test points in function approximation, and opponents in game-playing. Evaluation on all tests is often infeasible. Identification of an accurate evaluation or fitness function is a difficult problem in itself, and approximations are likely to introduce human biases into the search process. Coevolution {\em evolves} the set of tests used for evaluation, but has so far often led to inaccurate evaluation. We show that for any set of learners, a Complete Evaluation Set can be determined that provides ideal evaluation as specified by Evolutionary Multi-Objective Optimization. This provides a principled approach to evaluation in coevolution, and thereby brings {\em automatic} ideal evaluation within reach. The Complete Evaluation Set is of manageable size, and progress towards it can be accurately measured. Based on this observation, an algorithm named DELPHI is developed. The algorithm is tested on problems likely to permit progress on only a subset of the underlying objectives. Where all comparison methods result in overspecialization, the proposed method and a variant achieve sustained progress in all underlying objectives. These findings demonstrate that ideal evaluation may be approximated by practical algorithms, and that accurate evaluation for test-based problems is possible even when the underlying objectives of a problem are unknown. Any comments you may have on this work are very welcome. Best regards, Edwin de Jong -------------------------------------------------------------------- http://www.cs.uu.nl/~dejong From poznan at iub-psych.psych.indiana.edu Tue May 11 13:59:00 2004 From: poznan at iub-psych.psych.indiana.edu (Roman Poznanski) Date: Tue, 11 May 2004 12:59:00 -0500 Subject: Special Issue of JIN Dedicated to the Late Gen Matsumoto Message-ID: <40A11464.9000908@iub-psych.psych.indiana.edu> JOURNAL OF INTEGRATIVE NEUROSCIENCE VOLUME 3, ISSUE 2 (JUNE, 2004) Editorial: M. Ichikawa Prologue- the work of the late Professor Gen Matsumoto: K.Aihara, I.Tsuda, and H.Fujii Short Communications 1. I. TASAKI ON THE CONDUCTION VELOCITY OF NERVE FIBERS 2. M. ICHIKAWA and G. MATSUMOTO THE BRAIN-COMPUTER: ORIGIN OF THE IDEA AND PROGRESS IN ITS REALIZATION 3. H. TSUJINO OUTPUT-DRIVEN OPERATION AND MEMORY-BASED ARCHITECTURE PRINCIPLES EMBEDDED IN A REAL WORLD DEVICE. Research Reports 1. Y. YAMAGUCHI, Y.AOTA, N.SATO, H.WAGATSUMA, and Z.WU SYNCHRONIZATION OF NEURAL OSCILLATIONS AS A POSSIBLE MECHANISM UNDERLYING EPISODIC MEMORY: A STUDY OF THETA RHYTHM IN THE HIPPOCAMPUS 2. I. TSUDA, H.FUJII, S.TADOKORO, T. YASUOKA, and Y.YAMAGUTI CHAOTIC ITINERANCY AS A MECHANISM OF IRREGULAR CHANGES BETWEEN SYNCHRONIZATION AND DESYCHCHRONIZATION IN A NEURAL NETWORK. 3. H.FUJII, K. AIHARA, and I.TSUDA FUNCTIONAL RELEVANCE OF 'EXCITATORY' GABA ACTIONS IN CORTICAL INTERNEURONS: A DYNAMICAL SYSTEMS APPROACH. 4. S.DOI, J.INOUE, and S. KUMAGAI CHAOTIC SPIKING IN THE HODGKIN-HUXLEY NERVE MODEL WITH SLOW INACTIVATION OF THE SODIUM CURRENT 5. R.R. POZNANSKI ANALYTICAL SOLUTIONS OF THE FRANKENHAEUSER-HUXLEY EQUATIONS 6. P. BASSER SCALING LAWS FOR MYELINATED AXONS DERIVED FROM AN ELECTROTONIC CORE-CONDUCTOR MODEL ------------------------------------------------------------------------ For orders within Europe, please contact the Imperial College Press sales department at: Tel: +44 (0)20 7836-0888 Fax: +44 (0)20 7836-2020 during U.K. business hours. Outside Europe, our books and journals are distributed by World Scientific Publishing Co. World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, SINGAPORE 596224 Fax: 65-6467-7667 Tel: 65-6466-5775 E-mail: wspc at wspc.com.sg Price Information: ISSN: 0219-6352 ; Vol. 3/2004; 4 Issues Special Rates: Individuals -- Roman R. Poznanski, Ph.D Associate Editor, Journal of Integrative Neuroscience Department of Psychology Indiana University 1101 E. 10th St. Bloomington, IN 47405-7007 email: poznan at iub-psych.psych.indiana.edu phone (Office): (812) 856-7195 http://www.worldscinet.com/jin/mkt/editorial.shtml From masulli at disi.unige.it Tue May 11 04:57:08 2004 From: masulli at disi.unige.it (Francesco Masulli) Date: Tue, 11 May 2004 10:57:08 +0200 Subject: CFP: CIBB 2004 - INTERNATIONAL MEETING ON COMPUTATIONAL INTELLIGENCE METHODS FOR BIOINFORMATICS AND BIOSTATISTICS Message-ID: <200405111057.08616.masulli@disi.unige.it> CIBB 2004 INTERNATIONAL MEETING ON COMPUTATIONAL INTELLIGENCE METHODS FOR BIOINFORMATICS AND BIOSTATISTICS September 14-15, 2004, Perugia, ITALY CIBB 2004 runs as a special session of WIRN 04 - XV ITALIAN WORKSHOP ON NEURAL NETWORKS http://siren.dsi.unimi.it/conferences/SI/html/index.html FIRST CALL FOR PAPERS JOINTLY ORGANIZED BY - BIOPATTERNS - European Network of Excellence on Computational Intelligence for Biopattern analysis in Support of eHealthcare - SIREN - Italian Neural Networks Society AIMS CIBB 2004 addresses a cutting edge area of application of Neural Networks, Fuzzy Logic and Evolutionary Computation methods. Technical areas include, but are not limited to: - Data and methods for prognosis - Data and methods for diagnosis - Integration of clinical and genetic data - Proteomics - Pharmacogenetics CHAIRS Francesco Masulli, University of Pisa (Italy) Antonina Starita, University of Pisa (Italy) Roberto Tagliaferri, University of Salerno (Italy) IMPORTANT DATES Paper electronic submission deadline: June 15 2004 Notification of acceptance: July 15 2004 Submission of camera-ready papers: Sept 2 2004 SUBMISSIONS Papers must not be longer than 6 pages, including a cover sheet stating (1) Paper title; (2) Keywords; (3) Authors names and affiliations; (4) Corresponding author's name and contact details, including telephone/fax numbers and e-mail address. For electronic submission refer to http://siren.dsi.unimi.it/conferences/SI/html/index.html All accepted papers submitted by registered participants to WIRN 2004 will be included in the proceedings book of WIRN 2004, that will be published by an international Publisher. REGISTRATION FEES The registration to WIRN 2004 includes the free participation at CIBB 2004. ------------------------------------------------------- From terry at salk.edu Wed May 12 14:29:23 2004 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 12 May 2004 11:29:23 -0700 (PDT) Subject: NEURAL COMPUTATION 16:6 In-Reply-To: <200404140447.i3E4leM16495@dax.salk.edu> Message-ID: <200405121829.i4CITNt96552@dax.snl.salk.edu> Neural Computation - Contents - Volume 16, Number 6 - June 1, 2004 LETTERS Nonlinear Population Codes Maoz Shamir and Haim Sompolinsky Enhancement of Information Transmission Efficiency by Synaptic Failures Mark S. Goldman An Algorithm for the Detection of Faces on the Basis of Gabor Features and Information Maximization Hitoshi Imaoka and Kenji Okajima Analysis of Sparse Representation and Blind Source Separation Yuanqing Li, Andrzej Cichocki and Shun-ichi Amari Minimax Mutual Information Approach for Independent Component Analysis Deniz Erdogmus, Kenneth E. Hild II, Yadunandana N. Rao and Jose C. Principe Improving Generalization Capabilities of Dynamic Neural Networks Miroslaw Galicki, Lutz Leistritz , Ernst Bernhard Zwick and Herbert Witte A Modified Algorithm for Generalized Discriminant Analysis Wenming Zheng, Li Zhao and Ciarong Zou Stability-Based Validation of Clustering Solutions Tilman Lange, Volker Roth, Mikio L. Braun and Joachim M. Buhmann ----- 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 dimi at ci.tuwien.ac.at Wed May 12 07:51:32 2004 From: dimi at ci.tuwien.ac.at (Evgenia Dimitriadou) Date: Wed, 12 May 2004 13:51:32 +0200 (CEST) Subject: CI BibTeX Collection -- Update Message-ID: The following volumes have been added to the collection of BibTeX files maintained by the Vienna Center for Computational Intelligence: IEEE Transactions on Evolutionary Computation, Volumes 7/5 - 8/2 IEEE Transactions on Fuzzy Systems, Volumes 11/5 - 12/2 IEEE Transactions on Neural Networks, Volumes 14/5 - 15/3 Machine Learning, Volumes 53/3 - 55/2 Neural Computation, Volumes 15/10 - 16/5 Neural Networks, Volumes 16/7 - 17/4 Neural Processing Letters, Volumes 17/3 - 19/2 Journal of Statistical Software Volumes 1/1 - 10/5 Most files have been converted automatically from various source formats, please report any bugs you find. The complete collection can be downloaded from http://www.ci.tuwien.ac.at/services/BibTeX.html ftp://ftp.ci.tuwien.ac.at/pub/texmf/bibtex/ Best, Vivi ************************************************************************ * Evgenia Dimitriadou * ************************************************************************ * Institut fuer Statistik * Tel: (+43 1) 58801 10773 * * Technische Universitaet Wien * Fax: (+43 1) 58801 10798 * * Wiedner Hauptstr. 8-10/1071 * Evgenia.Dimitriadou at ci.tuwien.ac.at * * A-1040 Wien, Austria * http://www.ci.tuwien.ac.at/~dimi* ************************************************************************ From qian at brahms.cpmc.columbia.edu Thu May 13 18:53:22 2004 From: qian at brahms.cpmc.columbia.edu (Ning Qian) Date: Thu, 13 May 2004 18:53:22 -0400 Subject: papers on stereo and motor modeling available Message-ID: <40A3FC62.8090701@brahms.cpmc.columbia.edu> Dear Colleagues, The pdf files of two preprints from my lab are available at the links below. The first paper is about a multi-scale and multi-orientation stereo model, and the second is on comparing models of motor planning. The pdf files of some old papers, including the one on protein structure prediction that several people asked me for, can also be find at the same site http://brahms.cpmc.columbia.edu . Best regards, Ning ----------------------------------------------------------------- A Coarse-to-fine Disparity Energy Model with both Phase-shift and Position-shift Receptive Field Mechanisms Yuzhi Chen and Ning Qian, Neural Computation, 2004, in press. Numerous studies suggest that the visual system uses both phase- and position-shift receptive field (RF) mechanisms for the processing of binocular disparity. Although the difference between these two mechanisms has been analyzed before, previous work mainly focused on disparity tuning curves instead of population responses. However, tuning curve and population response can exhibit different characteristics, and it is the latter that determines disparity estimation. Here we demonstrate, in the framework of the disparity energy model, that for relatively small disparities, the population response generated by the phase-shift mechanism is more reliable than that generated by the position-shift mechanism. This is true over a wide range of parameters including the RF orientation. Since the phase model has its own drawbacks of underestimating large stimulus disparity and covering only a restricted range of disparity at a given scale, we propose a coarse-to-fine algorithm for disparity computation with a hybrid of phase-shift and position-shift components. In this algorithm, disparity at each scale is always estimated by the phase-shift mechanism to take advantage of its higher reliability. Since the phase based estimation is most accurate at the smallest scale when the disparity is correspondingly small, the algorithm iteratively reduces the input disparity from coarse to fine scales by introducing a {\em constant} position-shift component to all cells for a given location in order to offset the stimulus disparity at that location. The model also incorporates orientation pooling and spatial pooling to further enhance reliability. We have tested the algorithm on both synthetic and natural stereo images, and found that it often performs better than a simple scale averaging procedure. http://brahms.cpmc.columbia.edu/publications/stereo-scale.pdf Different Predictions by the Minimum Variance and Minimum Torque-Change Models on the Skewness of Movement Velocity Profiles Hirokazu Tanaka, Meihua Tai and Ning Qian, Neural Computation, 2004, in press. We investigated the differences between two well-known optimization principles for understanding movement planning: the minimum variance (MV) model of Harris \& Wolpert and the minimum torque-change (MTC) model of Uno et al. Both models accurately describe the properties of human reaching movements in ordinary situations (e.g. nearly straight paths and bell-shaped velocity profiles). However, we found that the two models can make very different predictions when external forces are applied or when the movement duration is increased. We considered a second-order linear system for the motor plant that has been used previously to simulate eye movements and single-joint arm movements, and were able to derive analytical solutions based on the MV and MTC assumptions. With the linear plant, the MTC model predicts that the movement velocity profile should always be symmetric, independent of the external forces and movement duration. In contrast, the MV model strongly depends on the movement duration and the system's degree of stability; the latter in turn depends on the total forces. The MV model thus predicts a skewed velocity profile under many circumstances. For example, it predicts that the peak location should be skewed toward the end of the movement when the movement duration is increased in the absence of any elastic force. It also predicts that with appropriate viscous and elastic forces applied to increase the system stability, the velocity profile should be skewed toward the beginning of the movement. The velocity profiles predicted by the MV model can even show oscillations when the plant becomes highly oscillatory. Our analytical and simulation results suggest specific experiments for testing the validity of the two models. http://brahms.cpmc.columbia.edu/publications/motor-models.pdf -- Ning Qian, Ph. D. Associate Professor Ctr. Neurobiology & Behavior Columbia University / NYSPI Kolb Annex, Rm 519 1051 Riverside Drive, Box 87 New York, NY 10032, USA http://brahms.cpmc.columbia.edu nq6 at columbia.edu 212-543-6931 ext 600 (Office) 212-543-5816 (Fax) From l.wiskott at biologie.hu-berlin.de Thu May 13 06:09:14 2004 From: l.wiskott at biologie.hu-berlin.de (Laurenz Wiskott) Date: Thu, 13 May 2004 12:09:14 +0200 Subject: PhD position available in Berlin Message-ID: <16547.18762.414079.62270@huxley.biologie.hu-berlin.de> Please forward this job advertisement to students who might be interested. Thanks, Laurenz Wiskott. ___________________________________________________________________________ Open Position for a PhD-Student in Computational Neuroscience at the Institute for Theoretical Biology, Berlin http://itb.biologie.hu-berlin.de/~wiskott/jobs.html ___________________________________________________________________________ Institute: Institute for Theoretical Biology Humboldt-Universit?t zu Berlin Invalidenstra?e 43 D-10115 Berlin, Germany http://itb.biologie.hu-berlin.de/ The Institute for Theoretical Biology is a young and dynamic lab with three full professors, three junior research groups, and about 60 students and researchers doing interdisciplinary and innovative research in different areas of theoretical biology. Research group: The position is available in the junior research group led by Dr. Laurenz Wiskott and funded by the Volkswagen Foundation. Research topics: The central issue addressed by the junior research group is the question of how invariant object perception can be achieved and learned in sensory systems, especially in the primate visual system. The project for this PhD position will be a theoretical analysis of simulation results. Teaching: There will be some duties as a teaching assistant, e.g. for the basic mathematics courses. Time: The position is available now. The initial appointment will be for 1 year with a possible extension by additional 2 years. Requirements: Candidates should have a recent degree (Diploma/MSc) in mathematics or theoretical physics. Required are strong mathematical and programming skills as well as the ability to communicate and work well in a team. Salary: Salary will be ? BAT IIa-O and will depend on age, and family status. BAT is the regular salary scale for public employees in Germany. Inquiries: Informal inquiries can be addressed to Dr. Laurenz Wiskott. Application: Complete applications should be sent to Dr. Laurenz Wiskott at the address given above. Please send only copies and not original documents, since the applications will not be sent back. You can also send applications via email, but please make sure they are complete and in a convenient format. Handicapped applicants with corresponding qualifications will be considered preferentially. To increase the proportion of female scientists, applications of qualified females are especially welcome. Deadline: None. Applications will be accepted until the position is filled, as will be indicated on the web-page of this job advertisement; see below. WWW: The following web page contains additional information: http://itb.biologie.hu-berlin.de/~wiskott/jobs.html. From C.Campbell at bristol.ac.uk Thu May 13 07:33:48 2004 From: C.Campbell at bristol.ac.uk (ICG Campbell, Engineering Mathematics) Date: Thu, 13 May 2004 12:33:48 +0100 Subject: RA position (Bristol University, UK) Message-ID: <562789.1084451628@ems-iggs.enm.bris.ac.uk> Research Assistant Position: Analysis of Microarray Data Applications are invited from experienced postdoctoral Research Fellows to work on a project principally focussed on new methods for the analysis and interpretation of microarray datasets. This project will involve the design of new algorithms and data analysis techniques to improve inference in medical diagnosis, predicted prognosis and fusion of data from different sources. We will use a variety of methods including Bayesian techniques and graphical methods, kernel methods and other approaches. This project is part of a large collaborative European Framework 6 consortium and will involve interaction and visits to collaborating groups in the UK and visits to overseas groups in Norway, France, Israel and the US. On the experimental side our collaborators are medical groups generating data for cancer research and diseases of the brain and the Research Fellow should be willing to sometimes assist in the interpretation and analysis of datasets generated by our collaborators. The Research Fellow will be hosted by the Computational Intelligence subgroup of the Artificial Intelligence Group within the Department of Engineering Mathematics, Bristol University. Details about our group and publications can be obtained from our website http://www.enm.bris.ac.uk/ai/ Applicants should have excellent mathematical and computational skills, and preferably a background in machine learning, statistics or bioinformatics. Details: Duration: 2 years with possible extension. Staff category and grade: RA1A Salary range: ?18893 - ?28279 per annum depending on qualifications and experience. Application: Informal enquiries should be made to Dr. Colin Campbell, email: C.Campbell(at)bris.ac.uk Formal applications and details of how to apply are available from the Personnel Services website, at http://www.bris.ac.uk/boris/jobs/ (vacancy ref. 10256). The closing date for applications is 30th June 2004. ---------------------- ICG Campbell, Engineering Mathematics C.Campbell at bristol.ac.uk From stefan.wermter at sunderland.ac.uk Fri May 14 06:05:57 2004 From: stefan.wermter at sunderland.ac.uk (Stefan Wermter) Date: Fri, 14 May 2004 11:05:57 +0100 Subject: NeuroBotics - Bioinspired computation for Robotics - Call for papers Message-ID: <40A49A05.1050305@sunderland.ac.uk> *************************************** International Workshop on NeuroBotics: Bioinspired Computation for Robotics **************************************** 20 September 2004 Call for Papers --------------- Substantial progress has been made recently in bio-inspired computation and robotics. This international workshop invites contributions to robotics which use methods of learning or artificial neural networks and/or are inspired by observations and results in neuroscience, cognitive science and animal behaviour. Topics of interest include but are not restricted to: ------- Neural networks for robots Biomimetic robots Learning for robotics Cognitive robots Speech interfaces and neural networks Neural Vision Talking robots Spiking neural networks in robots Learning self localization and mapping Imitation and neural networks Cognitive Development in robots etc Location -------- The workshop is organised as part of the 27th Conference on Artificial Intelligence (KI-2004) and runs parallel to the 34th Annual Meeting of the German Computer Science Society (Informatik 2004). The workshop will take place on September 20, 2004 at the University of Ulm, Germany. Format ------ We invite contributions for half-hour talks and, if possible, additional short demonstrations. If we receive sufficient submissions of high quality, we plan to publish the revised articles of the workshop contributions in a journal or book. Deadlines --------- Please send either abstracts of one page or full papers up to 8 pages (and if appropriate, possible descriptions of demonstrations) until June 9, 2004 to the organisers email address below. You will receive notification of acceptance, a more detailed workshop program and regarding the planned publication by July 5, 2004. Workshop registration: until August 20, 2004. Conference fees: will be not more than 100 Euro for the workshop. Registration at the conference will be optional. Organisation and more details: ------------------------------ More details about the AI-conference can be found at: http://ki2004.uni-ulm.de and more details about the organsing MirrorBot project to found at http://www.his.sunderland.ac.uk/mirrorbot/ and updates on this call at http://www.his.sunderland.ac.uk/mirrorbot/call.html Organisers ---------- Prof. Gnther Palm Neural Information Processing Computer Science Faculty University of Ulm Oberer Eselsberg D-89069 Ulm Germany palm at neuro.informatik.uni-ulm.de Prof. Stefan Wermter Hybrid Intelligent Systems School of Computing and Technology University of Sunderland St Peter's Way Sunderland SR6 0DD United Kingdom Stefan.Wermter at sunderland.ac.uk http://www.his.sunderland.ac.uk/~cs0stw/ From codrina.lauth at ais.fraunhofer.de Thu May 13 06:25:11 2004 From: codrina.lauth at ais.fraunhofer.de (Codrina Lauth) Date: Thu, 13 May 2004 12:25:11 +0200 Subject: Call for Demo Sessionat PKDD/ECML 2004: DEADLINE: 15 May 2004 Message-ID: <003d01c438d4$b035baf0$b4921a81@ais.fhg.de> Call for Demo Session The 15th European Conference on Machine Learning (ECML) and the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD) will be co-located in Pisa, Italy, September 20-24, 2004. The combined event will comprise presentations of contributed papers and invited speakers, a wide program of workshops, tutorials, and a discovery challenge, and last but not least a demo session. The Demo Session is reserved to innovative applications or successful experiences/prototypes in applying the Knowledge Discovery Process to practical cases. Also, are solicited demonstrations of prototypes/systems which show the integration of new Data Mining and Machine Learning technologies, advances using complex data structures and databases, XML-data, multimedia, expert systems in innovative application domains, such as bioinformatics, medical domains, WEB, e-commerce, etc. Demonstration proposals should give a short description of the demonstrated system, explain what is going to be demonstrated, and state the significance of the contribution to the field of Machine Learning and Knowledge Discovery in Databases. Important for the participants, in addition to the description of technical functionalities, will be to provide also performance results and quality measures (on standard benchmarks in the field and comparative examples). Short papers describing demos will appear in the Proceedings. Demonstration proposals must be no more than three (3) pages, in English and should be formatted according to the Springer-Verlag Lecture Notes in Computer Science guidelines. Authors instructions and style files can be downloaded at http://www.springer.de/comp/lncs/authors.html. Important Dates Submission Proposal deadline: Saturday May 15, 2004 Notification of acceptance: Monday June 7, 2004 Camera-ready copies due: Monday June 28, 2004 (Strict deadline) In the exhibit space, exhibitors give demonstrations and present their prototypes/systems to the audience, answering to their questions. A selected demo is shown to the interested audience in a slot of two hours including a coffee break. Each demo is then repeated more times (till three) during all week. Demos are explicitely mentioned in the conference program scheduling and in the daily news. Exhibitors are responsible for bringing all hardware and software required for their demonstrations. Accepted exhibitors have to register to the conference like all the other conference participants. Please submit your proposal to: KDNet Office Ina Lauth ------------------------------------------------------ Codrina Lauth MA Fraunhofer Institute for Autonomous Intelligent Systems (AIS), KD - Knowledge Discovery Team - Schloss Birlinghoven D-53754 Sankt Augustin, Germany Tel.: +49 - (0) 22 41 - 14 - 2686 Fax: +49 - (0) 22 41 - 14 - 2072 Mobile:+49 - (0)174 - 64 96 833 mailto:codrina.lauth at ais.fhg.de URL: http://www.ais.fraunhofer.de/KD/lauth.htm Institute: http://www.ais.fraunhofer.de Project: http://www.kdnet.org From mcgl-ci0 at wpmail.paisley.ac.uk Sat May 15 07:45:21 2004 From: mcgl-ci0 at wpmail.paisley.ac.uk (Stephen McGlinchey) Date: Sat, 15 May 2004 12:45:21 +0100 Subject: CFP - Special Session on Neural Networks in Games and Simulations Message-ID: C A L L F O R P A P E R S Special Session On NEURAL NETWORKS IN COMPUTER GAMES AND SIMULATION as part of International Conference on Computer Games: Artificial Intelligence, Design and Education (CGAIDE) 8-10 November 2004 Hosted at the Microsoft Campus, Reading, UK http://www.scit.wlv.ac.uk/~cm1822/cgaide.htm The main objectives of this special session are: * To provide a forum for presenting and discussing the application of neural networks in computer games and simulations * To promote collaboration between the computer games and computational intelligence research communities. Neural Networks (NN) in in Computer Games and Simulations ============================We invite submission of papers presenting recent research on Artificial Neural Networks applied to any aspect of computer games technology or simulation. The scope of the session may include topics from the following non-exclusive list. Autonomous control of game agents. Processing of game content data. Real-time feature extraction from game data. Simulation or anaylisis of emotions / intelligence. Automatic analysis of human player behaviours. Synthesis of dynamic game content. Strategic game applications (e.g. Go and Chess) Optimising search spaces. Submissions are encouraged from any other related application of neural networks in games or simulation. Full papers should be sent in Adobe pdf or MS Word format to stephen.mcglinchey at paisley.ac.uk by the submission deadline. Send a full paper - not previously published - in Word format (no smaller than 10-point format) to include author, address, e-mail, keywords and abstract (200 words max) in one of the following categories: Regular paper max. 5 pages Extended Paper max. 8 pages Critical Review paper max. 10 pages Important dates: Draft paper submission : 30 July 2004 Notification of acceptance : 23 August 2004 Camera-ready submission deadline : 13 September 2004 Please send enquiries to: Stephen McGlinchey (stephen.mcglinchey at paisley.ac.uk) School of Computing University of Paisley Paisley, Scotland or Darryl Charles (dk.charles at ulster.ac.uk) School of Computing and Information Engineering University of Ulster Coleraine, Northern Ireland. From Angela.Jakary at med.va.gov Sat May 15 18:48:26 2004 From: Angela.Jakary at med.va.gov (Angela.Jakary@med.va.gov) Date: Sat, 15 May 2004 17:48:26 -0500 Subject: Volunteer wanted to analyze psychiatric neuroimaging lab data. Po tential for NIH R21 grant collaboration! Message-ID: <6884EB498415D411A71A0000F803A4680A2C3A8E@VHASFCEXC1> Research Opportunity: Neural Networks Analysis of Neuroimaging Data Our psychiatric neuroimaging laboratory located in the San Francisco VA Medical Center is looking for an individual to apply neural network pattern recognition methods to our neuroimaging data (we have NeuralWare software but are open to suggestions). The ultimate goal is to discover new relationships between neuroimaging abnormalities--based on magnetic resonance imaging (MRI) tissue volume and proton magnetic resonance spectroscopic imaging (MRSI) metabolite measurements--and demographic/clinical data for individuals suffering from bipolar disorder, depression, or schizophrenia. It is hypothesized that the resultant relationships could suggest new, more accurate classification systems for these disorders. Knowledge of quantitative MRI and MRSI techniques is ideal, however training is available. This is an outstanding research opportunity for the right individual, and could evolve into grant-funded research. Specifically, the NIMH is interested in seeing new methods of data analysis applied to neuroimaging data, and we would therefore like to develop a competitive R21 grant proposal (no pilot data required) involving neural networks analysis of our data. If you're interested, please send an e-mail to angela.jakary at med.va.gov. Thanks. From matsuka at psychology.rutgers.edu Wed May 19 20:53:41 2004 From: matsuka at psychology.rutgers.edu (Matsuka, Toshihiko) Date: Wed, 19 May 2004 20:53:41 -0400 Subject: [Paper available] COMBINATORIAL CODES in Obj Recognition Message-ID: <004e01c43e04$e5149860$0d5fe6a5@mp> Dear Connectionists: We would like to announce the paper: COMBINATORIAL CODES IN VENTRAL TEMPORAL LOBE FOR OBJECT RECOGNITION: HAXBY (2001) REVISITED: IS THERE A "FACE" AREA? by Stephen Jose Hanson, Toshihiko Matsuka, and James V. Haxby. to appear in NeuroImage ---abstract--- Haxby et al (2001) recently argued that category-related responses in the ventral temporal (VT) lobe during visual object identification were overlapping and distributed in topography. This observation contrasts with prevailing views that object codes are focal and localized to specific areas such as the fusiform and parahippocampal gyri. We provide a critical test of Haxby's hypothesis using a Neural Network classifier which can detect more general topographic representations and achieves 83% correct generalization performance on patterns of voxel responses in out-of-sample tests. Using voxel-wise sensitivity analysis we show that substantially the same VT lobe voxels contribute to the classification of all object categories, suggesting the code is combinatorial. Moreover, we found no evidence for local single category representations. The neural network representations of the voxel codes were sensitive to both category and to superordinate level features which were only available implicitly in the object categories. A preprint is available at: http://www.rumba.rutgers.edu/pubs/objrecog_shj_tm_jvh.PDF Best regards, Toshihiko Matsuka ------------------ toshihiko matsuka rutgers university ------------------ From b.graham at cs.stir.ac.uk Thu May 20 12:46:14 2004 From: b.graham at cs.stir.ac.uk (Bruce Philip Graham) Date: Thu, 20 May 2004 17:46:14 +0100 (BST) Subject: FINAL CALL: Edinburgh Summer School in Neuroinformatics Simulation Tools] Message-ID: <2592.139.153.254.219.1085071574.squirrel@yen.cs.stir.ac.uk> 2004 EDINBURGH SUMMER SCHOOL IN NEUROINFORMATICS SIMULATION TOOLS Institute for Adaptive and Neural Computation University of Edinburgh. August 23-27, 2004 FINAL CALL: * Application deadline: 4th June 2004 * This summer school in Neuroinformatics Simulation Tools is a five day intensive course which will provide a practical introduction to using neuroscience simulation tools including NEURON and Catacomb for computational modelling of neural systems. The course will be held at the e-Science Institute in the centre of Edinburgh and will include talks from invited speakers as well as practical hands-on experience using the latest simulation tools. The course is aimed at PhD students, postdocs and faculty in neuroscience and related disciplines wishing to learn how to apply computational modelling techniques to their research problems. Bursaries are available for PhD students to help with travel and accommodation costs. For more details, and an online application form visit:- http://www.anc.ed.ac.uk/school -- Dr Bruce Graham, Lecturer (b.graham at cs.stir.ac.uk) Dept. of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA phone: +44 1786 467 432 fax: +44 1786 464 551 -- The University of Stirling is a university established in Scotland by charter at Stirling, FK9 4LA. Privileged/Confidential Information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not disclose, copy or deliver this message to anyone and any action taken or omitted to be taken in reliance on it, is prohibited and may be unlawful. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. From ltbip at gatsby.ucl.ac.uk Fri May 21 09:32:40 2004 From: ltbip at gatsby.ucl.ac.uk (ltbip@gatsby.ucl.ac.uk) Date: Fri, 21 May 2004 14:32:40 +0100 Subject: Workshop on Learning Theoretic and Bayesian Inductive Principles Message-ID: <200405211332.i4LDWekX021180@wald.gatsby.ucl.ac.uk> Call for Participation EU PASCAL Workshop on Learning Theoretic and Bayesian Inductive Principles 19-21 July 2004 Gatsby Unit University College London, UK http://www.gatsby.ucl.ac.uk/~zoubin/LTBIP The theoretical analysis of systems that learn from data has been an important topic of study in statistics, machine learning, and information theory. Formal tools for this analysis have been developed from distinct inductive paradigms, such as Bayesian inference, statistical learning theory, and minimum description length. In recent years, there has been a convergence of ideas from these distinct paradigms, an example of which are PAC-Bayesian bounds on generalisation performance. The goal of this workshop is to bring together leading theoreticians to allow them to debate, compare and cross-fertilise ideas from these distinct inductive principles. The format of the workshop will be 90 minute talks from invited speakers, short talks from contributed speakers, and ample discussion time. We invite original contributions to the workshop that will undergo a short refereeing process. Please describe the work in 1-4 pages specifying the model that you adopt and outlining the results/contribution. We encourage blue-skies ideas but they should be sufficiently concrete to be clear and understandable. The deadline for contributions is 31 May 2004. IMPORTANT DATES: Deadline for Contributions: 31 May 2004 Notification of Acceptance: 14 Jun 2004 Registration Deadline(*): 28 Jun 2004 Workshop Dates: 19-21 Jul 2004 INVITED SPEAKERS: Peter Bartlett, University of California at Berkeley, USA Olivier Catoni, Universite Paris 6, France A Philip Dawid, University College London, UK Claudio Gentile, Universita dell'Insubria, Italy Gabor Lugosi, Pompeu Fabra University, Spain John Langford, Toyota Technological Institute, USA David MacKay, University of Cambridge, UK Paul Vitanyi, CWI, Netherlands ORGANISERS: Zoubin Ghahramani, University College London, UK John Shawe-Taylor, University of Southampton, UK PROGRAMME COMMITTEE: Zoubin Ghahramani, University College London, UK Peter Grnwald, CWI, Netherlands John Langford, Toyota Technological Institute, USA Gabor Lugosi, Pompeu Fabra University, Spain Shahar Mendelson, Australian National University, Australia John Shawe-Taylor, University of Southampton, UK ---------------------------------------------------------------------- PASCAL (Pattern Analysis, Statistical Modelling and Computational Learning, http:///www.pascal-network.org) is a newly launched (December 2003) European Network of Excellence (NoE) as part of its IST program. The NoE brings together experts from basic research areas such as Statistics, Optimisation and Computational Learning and from a number of application areas, with the objective of integrating research agendas and improving the state of the art in all concerned fields with emphasis on applications in multi-modal interfaces. From chiba at Cogsci.ucsd.edu Sat May 22 10:00:44 2004 From: chiba at Cogsci.ucsd.edu (Andrea Chiba) Date: Sat, 22 May 2004 07:00:44 -0700 (PDT) Subject: ICDL 04 Deadline Extended May 23 In-Reply-To: Message-ID: Due to unexpected problems with the submission site, papers and abstracts are still being accepted through Sunday, May 23rd. > ICDL 2004 FINAL Call for papers > DEADLINE IS SUNDAY, MAY 23 2004 > There will be absolutely no further extensions. > > THIRD INTERNATIONAL CONFERENCE ON DEVELOPMENT AND LEARNING: DEVELOPING > SOCIAL BRAINS > > The Salk Institute > October 20-22, 2004 > San Diego, California > A Satellite Conference preceding > The Annual Society for Neuroscience Conference > http://www.icdl.cc > > The goal of the conference is to bring together leading researchers in > neuroscience, machine learning, robotics, and developmental psychology, in > order to gain new insights about learning and development in natural > organisms and robots. The scope of developmental processes to be > considered is broad, including cognitive, social, emotional, and many > other skills exhibited by humans, and animals. The theme of the conference > this year will be "Developing Social Brains", but other topics related to > development and learning are welcome. > > PAPER SUBMISSION > The extended submission deadline is May 21, 2004. Papers for the meeting > can be submitted ONLY through the conference's web site at: > http://www.icdl.cc. Papers can be submitted either as a 200 word summary > or as a full paper (max 8 typeset pages). > IMPORTANT: There will be NO further submission deadline extensions. > > SPECIAL ISSUE ON NEUROCOMPUTING > Authors of selected papers will be invited to submit an extended version > of their paper for publication in a special issue of the Neurocomputing > Journal, on Development, Learning, and the Social Brain, published by > Elsevier Science B.V. > (http://www.elsevier.nl/locate/neucom) > > INVITED TALKS (Not yet confirmed) > John Allman > Dana Ballard > Rodney Brooks > Eric Courchesne > Peter Dayan > Jeff Elman > William Greenough > James L. McClelland > Pietro Perona > Terrence Sejnowski > Joan Stiles > John Watson > > REVIEW PROCESS > All submitted papers will be reviewed by the program committee. Papers > will be judged and accepted for the meeting based on the clarity with > which the work is described and the relevance to the goals of the > conference. All accepted papers not selected for oral talks as well as > papers explicitly submitted as poster presentations will be included in > one of three evening poster sessions. Authors will be notified of the > presentation format of their papers by the beginning of August. > > ORGANIZING COMMITTEE: > General Chair: Javier R. Movellan: > Co-Chairs: Andrea Chiba, Gedeon Deak, Jochen Triesch. > Program Chair: Jochen Triesch and Tony Jebara. > Program Co-Chairs: Marian Stewart-Bartlett, Gwen Ford Littlewort. > Publications Chair: Gedeon Deak. > > ADVISORY BOARD: > Jeff Elman > Pat Langley > James L. McClelland > Sandy Pentland > Terrence Sejnowski > Mriganka Sur > Esther Thelen > Juyang Weng > > PROGRAM COMMITTEE: > Minoru Asada > Dana Ballard > Luis Baumela > Simon Baron-Cohen > Mark Baxter > Leslie Carver > Jeff Cohn > Kerstin Dautenhahn > Kenji Doya > Martha Farah > Teresa Farroni > Masahiro Fujita > Ann Graybiel > William Greenough > Michael Hasselmo > Shoji Itakura > Hiroshi Ishiguro > Robert Jacobs > David Kleinfeld > Mark Konishi > Kang Lee > Denis Mareshal > Risto Miikulainen > Douglas Nitz > Roz Picard > Steven Quartz > Rajesh Rao > Matthew Schlesinger > Gregor Schoener > Geoffrey Schoenbaum > Linda Smith > Olaf Sporns > Luc Steels > Valerie Stone > Manuela Veloso > Paul Verschure > Christoph von der Malsburg > Hiroyuki Yano > > CONFERENCE REGISTRATION: > Register online at http://www.icdl.cc > Student Registration is: $150 > Non-student Registration is: $290 > > > From tere-ic0 at wpmail.paisley.ac.uk Fri May 21 12:27:56 2004 From: tere-ic0 at wpmail.paisley.ac.uk (Valery Tereshko) Date: Fri, 21 May 2004 17:27:56 +0100 Subject: PhD studentship Message-ID: Research studentship to work on Emergent Collective Behaviour in Ensembles of Communicating Agents is available in School of Computing of the University of Paisley. The approach is based on consideration of an agent ensemble as dynamical system and studying the latter with the methods of nonlinear dynamics and statistical physics. Knowledge of basic nonlinear dynamics and good computational skill are essential. Experience in and knowledge of statistical physics, chaos theory, neural and biologically-inspired computations, and swarm intelligence are desirable. Contact: Valery Tereshko, School of Computing, University of Paisley, Paisley PA1 2BE. e-mail: valery.tereshko at paisley.ac.uk From b.graham at cs.stir.ac.uk Thu May 20 12:42:56 2004 From: b.graham at cs.stir.ac.uk (Bruce Philip Graham) Date: Thu, 20 May 2004 17:42:56 +0100 (BST) Subject: FINAL CALL: Edinburgh Summer School in Neuroinformatics Simulation Tools] Message-ID: <2574.139.153.254.219.1085071376.squirrel@yen.cs.stir.ac.uk> 2004 EDINBURGH SUMMER SCHOOL IN NEUROINFORMATICS SIMULATION TOOLS Institute for Adaptive and Neural Computation University of Edinburgh. August 23-27, 2004 FINAL CALL:* Application deadline: 4th June 2004 * This summer school in Neuroinformatics Simulation Tools is a five day intensive course which will provide a practical introduction to using neuroscience simulation tools including NEURON and Catacomb for computational modelling of neural systems. The course will be held at the e-Science Institute in the centre of Edinburgh and will include talks from invited speakers as well as practical hands-on experience using the latest simulation tools. The course is aimed at PhD students, postdocs and faculty in neuroscience and related disciplines wishing to learn how to apply computational modelling techniques to their research problems. Bursaries are available for PhD students to help with travel and accommodation costs. For more details, and an online application form visit:- http://www.anc.ed.ac.uk/school -- Dr Bruce Graham, Lecturer (b.graham at cs.stir.ac.uk) Dept. of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA phone: +44 1786 467 432 fax: +44 1786 464 551 From terry at salk.edu Mon May 24 18:48:33 2004 From: terry at salk.edu (Terry Sejnowski) Date: Mon, 24 May 2004 15:48:33 -0700 (PDT) Subject: NEURAL COMPUTATION 16:7 In-Reply-To: <200405121829.i4CITNt96552@dax.snl.salk.edu> Message-ID: <200405242248.i4OMmXL28618@dax.snl.salk.edu> Neural Computation - Contents - Volume 16, Number 7 - July 1, 2004 NOTES Bayesian Estimation of Stimulus Responses in Poisson Spike Trains Sidney R. Lehky Comments on "A Parallel Mixture of SVMs for Very Large Scale Problems" Xiaomei Liu, Lawrence O. Hall, and Kevin W. Bowyer LETTERS Automated Algorithms for Multiscale Morphometry of Neuronal Dendrites Christina M. Weaver, Patrick R. Hof, Susan L. Wearne, and W. Brent Lindquist Computing and Stability in Cortical Networks Peter E. Latham and Sheila Nirenberg Real-Time Computation at the Edge of Chaos in Recurrent Neural Networks Nils Bertschinger and Thomas Natschlaeger Information Geometry of U-Boost and Bregman Divergence Noboru Murata, Takashi Takenouchi, Takafumi Kanamori and Shinto Eguchi A New Approach to the Extraction of ANN Rules and Their Generalization Capacity Through GP Juan R. Rabunal, Julian Dorado, Alejandro Pazos, Javier Pereira and Daniel Rivero A Classification Paradigm for Distributed Vertically Partitioned Data Jayanta Basak and Ravi Kothari ----- 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 michael.spratling at kcl.ac.uk Tue May 25 04:54:29 2004 From: michael.spratling at kcl.ac.uk (Michael Spratling) Date: Tue, 25 May 2004 09:54:29 +0100 Subject: Post-doc position, King's College London Message-ID: <40B309C5.3060708@kcl.ac.uk> Postdoctoral Research Associate King's College London 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 23,144 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 auke.ijspeert at epfl.ch Tue May 25 08:09:13 2004 From: auke.ijspeert at epfl.ch (Auke Jan Ijspeert) Date: Tue, 25 May 2004 14:09:13 +0200 Subject: SAB2004: call for abstracts and early registration deadlines (May 28 and 31) Message-ID: <40B33769.3080709@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. Best regards, Auke Ijspeert ----- SAB2004 Call for participation, Call for Abstracts ----- 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 ----- We are glad to announce that the registration for SAB2004 is now open, and that the preliminary technical program is available. The deadline for early registration is 28 May 2004. See http://www.isab.org/sab04/ for details. ----- Call for abstracts ----- SAB2004 introduces a new feature: the Last-Minute-Results poster session. 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. Deadlines: May 31 2004: Abstract submission deadline June 7 2004: Notification of acceptance 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 From d.g.heinke at bham.ac.uk Tue May 25 11:33:09 2004 From: d.g.heinke at bham.ac.uk (Dietmar Heinke) Date: Tue, 25 May 2004 16:33:09 +0100 Subject: Position in Computational Neuroscience Message-ID: <40B36735.5000706@bham.ac.uk> Research Fellow/Assistant in Computational Neuroscience. Behavioural Brain Sciences Centre School of Psychology University of Birmingham Birmingham The Fellow/Assistant will work with Dr Dietmar Heinke and Prof Glyn Humphreys in developing a computational model of visual object recognition and attention in humans. The framework for this project is set by the computational model, SAIM (Selective Attention for Identification Model; see http://web.bham.ac.uk/heinkedg/SAIM for details) and will draw on experimental evidence from a broad range of sources: single cell studies, fMRI, MR tractography, neuropsychological and psychological studies. The position is funded by the BBSRC and is part of the collaboration of the following four groups: Glyn Humphreys and Dietmar Heinke, Behavioural Brain Sciences Centre, Birmingham; S. Shipp, Institute of Cognitive Neuroscience, University College London; Paul Matthews and colleagues, Oxford Centre for Functional Magnetic Resonance Imaging of the Brain and Andrew T. Smith, Royal Holloway & Bedford New College. The project is part of a long-term effort in the Behavioural Brain Sciences Centre (BBS) to establish computational modelling as a standard method within cognitive neuroscience. Examples for first results can be found on Dietmar Heinke's web-page and the BBS web-page. This long-term project is also supported by long-standing collaborations with the school of computer science, including Aaron Sloman, Ela Claridge and John Bullinaria Applicants should have, or be about to complete, a PhD or a very good MSc in a relevant field (incl. computer vision, neural networks, connectionist modelling, physics, or applied mathematics). Familiarity with programming (MatLab, C++) would be an advantage. The posts are available from 1 September 2004 or as soon as possible thereafter. Application forms are available from Personnel Services, University of Birmingham (j.s.thorp at bham.ac.uk) and informal queries should be addressed to Dietmar Heinke (0121 414 4920 d.g.heinke at bham.ac.uk). The salary is at point 6 on the RA1A scale (18,893 to 21,010). The deadline is 30 June 2004. -- School of Psychology University of Birmingham Birmingham B15 2TT, UK http://web.bham.ac.uk/heinkedg/ Phone: +44 121-414-4920 FAX: +44 121-414-4897 From geoff at cns.georgetown.edu Wed May 26 09:46:50 2004 From: geoff at cns.georgetown.edu (Geoff Goodhill) Date: Wed, 26 May 2004 09:46:50 -0400 Subject: Papers on axon guidance available Message-ID: Dear Connectionists, The following two recent papers may be of interest to readers of this list. They can be downloaded from http://cns.georgetown.edu/~geoff/pubs.html Rosoff, W.J., Urbach, J.S., Esrick, M., McAllister, R.G., Richards, L.J. & Goodhill, G.J. (2004). A new chemotaxis assay shows the extreme sensitivity of axons to molecular gradients. Nature Neuroscience, 7, 678-682. Goodhill, G.J., Gu, M. & Urbach, J.S. (2004). Predicting axonal response to molecular gradients with a computational model of filopodial dynamics. Neural Computation, in press. Abstracts are appended below. Sincerely, Geoff Geoffrey J Goodhill, PhD Department of Neuroscience Georgetown University Medical Center 3900 Reservoir Road NW, Washington DC 20007 Email: geoff at georgetown.edu Homepage: cns.georgetown.edu ------------ A new chemotaxis assay shows the extreme sensitivity of axons to molecular gradients. Axonal chemotaxis is believed to play a key role in wiring up the developing and regenerating nervous system, but little is known about how axons actually respond to molecular gradients. We report a new quantitative assay that allows the long-term response of axons to gradients of known and controllable shape to be examined in a three-dimensional gel. Using this assay we show that axons may be nature's most sensitive gradient detectors, but that this sensitivity exists only within a narrow range of ligand concentrations. This assay should also be applicable to other biological processes controlled by molecular gradients, such as cell migration and morphogenesis. ------------ Predicting axonal response to molecular gradients with a computational model of filopodial dynamics. Axons are often guided to their targets in the developing nervous system by attractive or repulsive molecular concentration gradients. We propose a computational model for gradient sensing and directed movement of the growth cone mediated by filopodia. We show that relatively simple mechanisms are sufficient to generate realistic trajectories for both the short term response of axons to steep gradients and the long term response of axons to shallow gradients. The model makes testable predictions for axonal response to attractive and repulsive gradients of different concentrations and steepness, the size of the intracellular amplification of the gradient signal, and the differences in intracellular signaling required for repulsive versus attractive turning. From J.L.Wyatt at cs.bham.ac.uk Wed May 26 09:56:21 2004 From: J.L.Wyatt at cs.bham.ac.uk (Jeremy L Wyatt) Date: Wed, 26 May 2004 14:56:21 +0100 (BST) Subject: looking for two outstanding post-docs for an ambitious EC cognitive systems project Message-ID: Dear colleague, I'm writing to a long list of people because we are looking for two outstanding postdocs to work on a large EU funded project called CoSy. The project involves seven sites around the EU, including people working in natural language, robotics, vision, learning, psychology and AI. The aim is to put together many things that are normally done separately, e.g. vision, language, planning, plan execution, 3-D manipulation, learning, self-understanding, etc. integrated in a working robot in a coherent architecture. The project is described in greater detail here: http://www.cs.bham.ac.uk/research/projects/cosy/ Because of the breadth of the project we hope to find people with broad and deep knowledge of AI, at least one of whom should have experience in robotics. Because the tasks are so challenging we have managed to obtain funding for higher salaries than usual (at least in the UK) for post-doctoral positions. The Birmingham portion of the project is led by Aaron Sloman and myself. If you know of any potential candidates please pass on the appended advertisement directly, or circulate within your group. If you are interested then some of the work in the Birmingham part of the project is described here: http://www.cs.bham.ac.uk/research/projects/cosy/PlayMate-start.html If you have time, please feel free to comment, criticise, join in, etc. We hope that the project will be as open as possible, including the sharing of code, as well as ideas. The advert follows. cheers Jeremy PS Aaron Sloman is also circulating the advert in a similar way. While we've tried to make our mailing lists disjoint, I'd like to apologise if you've seen this advert more than once. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Dr Jeremy L Wyatt Intelligent Robotics Laboratory Tel 00 44 121 414 4788 School of Computer Science Fax 00 44 121 414 4281 University of Birmingham jlw at cs.bham.ac.uk Birmingham, B15 2TT, UK www.cs.bham.ac.uk/~jlw www.cs.bham.ac.uk/research/robotics -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ________________ Subject: Two Research Posts Birmingham UK: EC Cognitive Systems Robotic project THE UNIVERSITY OF BIRMINGHAM SCHOOL OF COMPUTER SCIENCE http://www.cs.bham.ac.uk/ Research Fellows (2 Posts) on EC-Funded 'Cognitive Systems' Integrated Project Applications are invited for two experienced researchers to work on an ambitious seven partner EC-funded project (CoSy: Cognitive Systems for Cognitive Assistants). The whole project will involve upward of twenty people including partners working in robotics, natural language, experimental and developmental psychology, computer vision, machine learning and artificial intelligence. The work at Birmingham will involve designing and implementing mechanisms and a cognitive architecture for a robot that sees, manipulates and talks about objects on a table. Much of the emphasis at Birmingham will be on developing and implementing architectures that allow us to integrate the various components of an intelligent system. The aim of the whole project is to investigate ways of developing physically embodied robots that combine perception, learning, reasoning, planning, acting, communication and self-awareness in an integrated architecture. One of the posts will require some robotics experience. The remit of the project is, however, very broad so people with experience in any area of AI are encouraged to apply. Relevant areas include, but are not limited to: reasoning, representation, architectures, learning, natural language understanding, robotics, and computer vision. Further details of the project and a draft description of work to be done in Birmingham can be found at: http://www.cs.bham.ac.uk/research/projects/cosy/ A job specification with instructions for applicants is available at http://www.cs.bham.ac.uk/news/jobs/cosy.rf.04/ Informal enquiries can be made to the two Principal Investigators at Birmingham: Dr Jeremy Wyatt or Professor Aaron Sloman Starting salary: on scale 21,852 - 31,715 a year depending on experience and qualifications, with potential progression to 34,838 (Scales are currently under review). Likely starting date for the project is: 1 September 2004. Appointments will be for a period of 18 months initially, with an expected extension to 48 months (standard EC procedure). Application forms are returnable by 10th June 2004, though late applicants may be considered. Further enquiries and requests for application forms may be made to Personnel Services, The University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK. Tel: +44 121 415 9000 Email: Ms. Jo Gerald Web: www.punit.bham.ac.uk/vacancies Please quote reference S36732 in all communications. Working towards equal opportunities. ==== From oby at cs.tu-berlin.de Fri May 28 03:40:49 2004 From: oby at cs.tu-berlin.de (Klaus Obermayer) Date: Fri, 28 May 2004 09:40:49 +0200 (MEST) Subject: tenured faculty position Message-ID: Dear All, below please find an announcement for an open faculty position (tenured) in the area of artificial intelligence in our department (EE and CS) at the Technical University of Berlin. The Berlin area has a high concentration of high quality research institutions (three universities, Max-Planck and Fraunhofer institutes, etc.) and a lively machine learning and computational neuroscience scene. Besides that, it is a pleasant city to live in and Germany's most prominent place for cultural activities. Cheers Klaus ========================================================================== FACULTY POSITION IN ARTIFICIAL INTELLIGENCE Faculty IV, Technical University of Berlin, Berlin, Germany The Department for Electrical Engineering and Computer Science solicits application for a tenured faculty position (salary level C3) in the area of artificial intelligence. Potential areas of research include (but are not restricted to) machine learning, automatic deduction and problem solving, planning and decision making, knowledge representation, scene analysis, and speech processing. The successful candidate is expected to join the department's undergraduate teaching programs, as well as the graduate education in the area of Artificial Intelligence. The successful candidate is expected to teach courses in German after a few years. Requirements: Ph.D. degree and Habilitation or equivalent achievements (cf. =A7100 BerlHG); research experience in the field of artificial intelligence; a strong publication record and teaching experience. Experience in the acquisition of research grants is desirable. Please send applications to: Chair of the Department of Electrical Engineering and Computer Science Technical University of Berlin FR 5-1, Franklinstrasse 28/29 10587 Berlin, Germany email: fbv13 at cs.tu-berlin.de All applications received before June 25th, 2004, will be given full consideration. The Technical University of Berlin wants to increase the percentage of women on its faculty and strongly encourages applications from qualified individuals. Women will be preferred given equal qualifications. Handicapped persons will be preferred given equal qualifications. ========================================================================= Prof. Dr. Klaus Obermayer phone: 49-30-314-73442 FR2-1, NI, Informatik 49-30-314-73120 Technische Universitaet Berlin fax: 49-30-314-73121 Franklinstrasse 28/29 e-mail: oby at cs.tu-berlin.de 10587 Berlin, Germany http://ni.cs.tu-berlin.de/ From zhong at cse.fau.edu Fri May 28 15:32:19 2004 From: zhong at cse.fau.edu (zhong@cse.fau.edu) Date: Fri, 28 May 2004 15:32:19 -0400 (EDT) Subject: CFP: ICTAI-2004 special track on Unsupervised and Semi-supervised Learning Message-ID: <36833.192.35.232.241.1085772739.squirrel@webmail.cse.fau.edu> (sincere apologies if you received multiple copies of this notice) Dear Colleagues, We would like to bring to your attention a special track of ICTAI-2004 on unsupervised and semi-supervised learning. The 16th International Conference on Tools with Artificial Intelligence (ICTAI-2004) will be held in Boca Raton, Florida on November 15-17, 2004. The special track will be part of the conference with a focus on reporting recent advances in theories and applciations of unsupervised and semi-supervised learning techniques. The web site for the main conference is http://www.cse.fau.edu/~ictai04/ and the URL for the CFP of the special track is http://www.cse.fau.edu/~zhong/cfp-learning.htm Deadlines: June 18 - paper submission due August 2 - acceptance notification September 3 - final camera ready paper due Any questions please email the co-chairs of this special track at davidson at cs.albany.edu or zhong at cse.fau.edu. Also please feel free to distribute this notice to people who you think are interested. sincerely, Ian Davidson Shi Zhong Co-chairs ICTAI-2004 Special Track on Theories and Applications of Unsupervised and Semi-Supervised Learning From ash at cs.umass.edu Fri May 28 10:32:00 2004 From: ash at cs.umass.edu (Ashvin Shah) Date: Fri, 28 May 2004 10:32:00 -0400 (EDT) Subject: links to J Neurophys article: Cortical Involvement in the Recruitment of Wrist Muscles In-Reply-To: Message-ID: Hello Connectionists - the following article may be of interest. I've included the citation and the abstract. Thank you. - Ashvin Shah, ash at cs.umass.edu, (413) 545-1596 Neuroscience and Behavior Program, University of Massachusetts, Amherst Shah, A., Fagg, A. H., and Barto, A. G. (2004) Cortical Involvement in the Recruitment of Wrist Muscles Journal of Neurophysiology vol. 99(6), pages 2445 - 2456 http://www-all.cs.umass.edu/pubs/2004/shah_fb_JNP04.pdf http://www-all.cs.umass.edu/pubs/2004/shah_fb_JNP04.ps http://jn.physiology.org/ PMID: 14749314 [PubMed - in process] ABSTRACT: In executing a voluntary movement, one is faced with the problem of translating a specification of the movement in task space (e.g., a visual goal) into a muscle-recruitment pattern. Among many brain regions, the primary motor cortex (MI) plays a prominent role in the specification of movements. In what coordinate frame MI represents movement has been a topic of considerable debate. In a two-dimensional wrist step-tracking experiment, Kakei et al. described some MI cells as encoding movement in a muscle-coordinate frame and other cells as encoding movement in an extrinsic-coordinate frame. This result was interpreted as evidence for a cascade of transformations within MI from an extrinsic representation of movement to a muscle-like representation. However, we present a model that demonstrates that, given a realistic extrinsic-like representation of movement, a simple linear network is capable of representing the transformation from an extrinsic space to the muscle-recruitment patterns implementing the movements on which Kakei et al. focused. This suggests that cells exhibiting extrinsic-like qualities can be involved in the direct recruitment of spinal motor neurons. These results call into question models that presume a serial cascade of transformations terminating with MI pyramidal tract neurons that vary their activation exclusively with muscle activity. Further analysis of the model shows that the correlation between the activity of an MI neuron and a muscle does not predict the strength of the connection between the MI neuron and muscle. This result cautions against the use of correlation methods as a measure of cellular connectivity. -- From poznan at iub-psych.psych.indiana.edu Fri May 28 15:43:36 2004 From: poznan at iub-psych.psych.indiana.edu (Roman R. Poznanski) Date: Fri, 28 May 2004 14:43:36 -0500 Subject: A new book from CRC Press on Neuromimetic robotics Message-ID: <40B79668.8080804@iub-psych.psych.indiana.edu> MODELING IN THE NEUROSCIENCES: FROM BIOLOGICAL SYSTEMS TO NEUROMIMETIC ROBOTICS SECOND EDITION Edited by G.N.Reeke, R.R. Poznanski, K.A. Lindsay, J.R. Rosenberg, and O.Sporns The Second Edition is a thoroughly updated and carefully written and edited volume, dedicated to covering recent conceptual advances in the neurosciences as well as the basics of the field. It is essential for those interested in theoretical neuroscience and its implementation via integrative and synthetic neuronal modeling. Many state-of-the-art examples are presented to show how mathematical and computer modeling can contribute at various levels in the hierarchy of the neurosciences. Each chapter includes a problem set. Data sets are available online to add coverage for future advances in this rapidly changing and expanding field. This seminal book develops models that provide insights into underlying physiological mechanisms at the molecular, cellular, and network levels. Physiological mechanisms, such as voltage-dependent ion channels, calcium-dependent release, neurotransmitter-receptor dynamics and second-messenger effects, backpropagating sodium spikes, as well as cable properties governing electric current flow in neurons and neuronal networks are discussed throughout the book. The final chapters outline directions of theoretical modeling leading towards neural robotics and the fabrication of neuromimetic robotic devices beyond what is currently available in today's robotic technology. Dimensions Hardcover (grey vinyl board with gold lettering on light green spine) + jacket cover (maroon red cover with golden green micrograph superimposed on black background), Approx. 740 pages, 280 illus., Bibliography :-1600 refs. Further information: www.crcpress.com . Sale Price: US$139.95 --------------070505030109010106050503 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit
MODELING IN THE NEUROSCIENCES: FROM BIOLOGICAL SYSTEMS TO NEUROMIMETIC ROBOTICS
SECOND EDITION

Edited by
G.N.Reeke, R.R. Poznanski, K.A. Lindsay, J.R. Rosenberg, and O.Sporns

The Second Edition is a thoroughly updated and carefully written and edited
volume, dedicated to covering recent conceptual advances in the neurosciences
as well as the basics of the field. It is essential for those interested in
theoretical neuroscience and its implementation via integrative and synthetic
neuronal modeling. Many state-of-the-art examples are presented to show how
mathematical and computer modeling can contribute at various levels in the
hierarchy of the neurosciences. Each chapter includes a problem set. Data
sets are available online to add coverage for future advances in this rapidly
changing and expanding field.
This seminal book develops models that provide insights into underlying
physiological mechanisms at the molecular, cellular, and network levels.
Physiological mechanisms, such as voltage-dependent ion channels,
calcium-dependent release, neurotransmitter-receptor dynamics and 
second-messenger effects, backpropagating sodium spikes, as well as cable
properties governing electric current flow in neurons and neuronal networks
are discussed throughout the book. The final chapters outline directions of
theoretical modeling leading towards neural robotics and the fabrication of
neuromimetic robotic devices beyond what is currently available in
today's robotic technology.

Dimensions

Hardcover (grey vinyl board with gold lettering on light green spine) +
jacket cover (maroon red cover with golden green micrograph superimposed on
black background), Approx. 740 pages, 280 illus., Bibliography :-1600 refs.

Further information:  www.crcpress.com.
Sale Price: US$139.95







--------------070505030109010106050503-- From d.mandic at imperial.ac.uk Fri May 28 07:04:54 2004 From: d.mandic at imperial.ac.uk (Danilo P. Mandic) Date: Fri, 28 May 2004 12:04:54 +0100 Subject: Signal characterisation - TR and article Message-ID: <037901c444a3$9b2f1a90$5a7dc69b@MandicLaptop> Dear Connectionists may I draw your attention to the Technical Report entitled: On the Characterisation of the Deterministic/Stochastic and Linear/Nonlinear Nature of Time Series (T. Gautama, M. M. Van Hulle and D. P. Mandic)=20 It can be downloaded from http://www.commsp.ee.ic.ac.uk/~mandic It combines the concept of surrogate data and signal representation in phase space to provide an insight into the nature of a signal. We have applied it to fMRI, HRV, and EEG signals and it works fine. A more rigorous analysis can be found in the following Physica D article: Temujin Gautama, Danilo P. Mandic and Marc M. Van Hulle ' The delay vector variance method for detecting determinism and nonlinearity in time series ' Physica D: Nonlinear Phenomena, 190 (2004) 167-176=20 which can be downloaded from: http://134.58.34.50/temu/physD04_preprint.pdf or http://www.commsp.ee.ic.ac.uk/~mandic=20 The reasoning behind this analysis is that the recent progress in biomedicine has brought to light problems where an insight into the nature of a signal is necessary before performing the actual signal processing. This particularly applies to the linear vs nonlinear and deterministic vs stochastic signal properties (remember the Wold decomposition theorem?) The MatLab code which accompanies the articles can be downloaded from: DVV-toolbox: http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=3264&objectType=file and Surrogates toolbox: http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=4612&objectType=FILE Your comments are most welcome. Enjoy! Danilo ============================= Dr Danilo Mandic Senior Lecturer in Signal Processing Department of Electrical and Electronic Engineering Imperial College London United Kingdom Phone: +44 207 594 6271 Fax: +44 207 594 6234 E-mail: d.mandic at imperial.ac.uk http://www.commsp.ee.ic.ac.uk/~mandic From nips04pub at hotmail.com Mon May 31 16:42:33 2004 From: nips04pub at hotmail.com (John Platt) Date: Mon, 31 May 2004 13:42:33 -0700 Subject: NIPS 2004 Call for Workshops Message-ID: CALL FOR WORKSHOPS --- NIPS 2004 Neural Information Processing Systems --- Natural and Synthetic Friday, December 17 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc Deadline for Workshop Proposals: August 1, 2004 Following the regular program of the Neural Information Processing Systems 2004 Conference in Vancouver, BC, Canada, workshops on various current topics in Neural Information Processing will be held on December 17 and 18, 2004, in Whistler, BC, Canada. We invite researchers interested in chairing one of these workshops to submit proposals for possible workshops. The goal of the workshops is to provide an informal forum for researchers to discuss important research questions and challenges. Controversial issues, open problems, and comparisons of competing approaches are encouraged and preferred as workshop topics. Representation of alternative viewpoints and panel-style discussions are also particularly encouraged. Workshop topics include, but are not limited to, the following: Active Learning, Attention, Audition, Bayesian Networks, Bayesian Statistics, Benchmarking, Bioinformatics, Brain Imaging, Computational Complexity, Control, Genetic Algorithms, Graphical Models, Hippocampus and Memory, Human-Computer Interfaces, Hybrid Supervised/Unsupervised Learning, Implementations, Independent Component Analysis, Kernel Methods, Mean-Field Methods, Music, Network Dynamics, Neural Coding, Neural Plasticity, On-Line Learning, Optimization, Recurrent Nets, Robotics, Rule Extraction, Self-Organization, Sensory Biophysics, Signal Processing, Spike Timing, Speech, Time Series, Topological Maps, and Vision. Detailed descriptions of previous workshops may be found at: http://nips.cc under Previous Conferences. There will be six hours of workshop meetings per day, split into morning and afternoon sessions, with free time between the sessions for ongoing individual exchange or outdoor activities. Selected workshops may be invited to submit proceedings for publication in the post-NIPS workshops monographs series published by the MIT Press. Workshop organizers have several responsibilities including: * Coordinating workshop participation and content, which includes arranging short informal presentations by experts, arranging for expert commentators to sit on a discussion panel, formulating a set of discussion topics, etc. * Moderating the discussion, and reporting its findings and conclusions to the group during the evening plenary sessions. * Writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. SUBMISSION INSTRUCTIONS: Interested parties must submit a proposal for a workshop via email by August 1,2004. Proposals should include a title, description of what the workshop is to address and accomplish, proposed workshop length (1 or 2 days), planned format (e.g., lectures, group discussions, panel discussion, combinations of the above, etc.), and proposed speakers. Names of potential invitees should be given where possible. Preference will be given to workshops that reserve a significant portion of time for open discussion or panel discussion, as opposed to a pure "mini-conference" format. An example format is: * Tutorial lecture providing background and introducing terminology relevant to the topic. * Discussion or panel presentation. * Short talks or panels alternating with discussion and question/answer sessions. * General discussion and wrap-up. We suggest that organizers allocate at least 50 percent of the workshop schedule to questions, discussion, and breaks. Past experience suggests that workshops otherwise degrade into mini-conferences as talks begin to run over. For the same reason, we strongly recommend that each workshop include no more than 12 talks per day. The proposal should state why the topic is of interest, why it should be discussed, and who the targeted group of participants is. It should also include a brief resume of the prospective workshop chair with a list of publications to establish scholarship in the field. We encourage workshops that build, continue, or arise from one or more workshops from previous years. Please mention any such connections. NIPS does not provide travel funding for workshop speakers. In the past, some workshops have sought and received funding from external sources to bring in outside speakers. In addition, the organizers of each accepted workshop can name up to four people (six people for 2-day workshops) to receive discounted registration for the workshop program. Submissions should include the name, address, email address, phone and fax numbers for all organizers. If there is more than one organizer, please designate one organizer as the primary contact. Proposals should be emailed as plain text to nips-prp at umich.edu (please do not use attachments, Word, postscript, html, or PDF files) NIPS 2004 WORKSHOP CO-CHAIRS: * Satinder Singh, University of Michigan * Daniel D. Lee, University of Pennsylvania Questions may be emailed to nips-adm at umich.edu DEADLINE FOR WORKSHOP PROPOSALS: AUGUST 1, 2004 REMINDER: Deadline for Papers --- June 4, 2004 From torsello at dsi.unive.it Mon May 31 08:59:37 2004 From: torsello at dsi.unive.it (Andrea Torsello) Date: Mon, 31 May 2004 14:59:37 +0200 Subject: Special Issue on Similarity-Based Pattern Recognition Message-ID: <200405311459.37577.torsello@dsi.unive.it> CALL FOR PAPERS PATTERN RECOGNITION Special Issue on Similarity-Based Pattern Recognition Submission deadline: December 15, 2004 Guest Editors: Manuele Bicego, University of Sassari, Italy (bicego at uniss.it) Vittorio Murino, University of Verona, Italy (vittorio.murino at univr.it) Marcello Pelillo, University of Venice, Italy (pelillo at dsi.unive.it) Andrea Torsello, University of Venice, Italy (atorsell at dsi.unive.it) Traditional Pattern Recognition techniques are centered around the notion of "feature". According to this view, the objects to be clustered or classified are represented in terms of properties that are intrinsic to the object itself. Hence, a typical pattern recognition system makes its "decisions" by simply looking at one or more feature vectors fed as input. The strength of this approach is that it can leverage a wide range of mathematical tools ranging from statistics, to geometry, to optimization techniques. However, in many real-world applications the objects are not naturally representable in terms of a vector of features. For example, graph representations lack a canonical order or correspondence between nodes. Furthermore, even if a vector mapping can be established, the vectors will be of variable length, hence would not belong to a single vector-space. On the other hand, quite often it is possible to obtain a measure of the similarity/dissimilarity of the objects to be classified. It is therefore tempting to design a pattern recognizer which, unlike traditional systems, accepts as input a matrix containing the similarities between objects and produces class labels as output. Indeed, recently, there has been a renewed interest in similarity-based techniques, both in developing and studying new effective distances between non vectorial entities, like graphs, sequences, structures, and in proposing alternative distance-based paradigms. These methods typically keep the algorithm generic and independent from the actual data representation, allowing the use of non-metric similarities (thereby violating the triangular inequality). Further, they make the approaches applicable to problems that do not have a natural embedding to a uniform feature space, such as the clustering of structural or graph-based representations or the analysis of sequences. Finally, these representations are well suited to both supervised and unsupervised classification. The literature of pairwise algorithms includes, among others, kernel methods, spectral clustering techniques/graph partitioning, quadratic optimization, self-organizing maps, etc. These techniques are successfully applied to very diverse problems like object classification, image retrieval by content, color quantization, image segmentation, perceptual grouping, and bioinformatics (gene alignment, gene classification or phylogenetic analysis). The goal of this special issue is to solicit and publish high-quality papers that bring a clear picture of the state of the art in this area. We aim to appeal to researchers in Pattern Recognition and Computer Vision who are using or developing similarity-based techniques. Papers are solicited that address theoretical as well as practical issues related to the Special Issue's theme. Topics of interest include (but are not limited to): - Similarity-based classification - Similarity-based clustering - Embeddings - Kernel methods - Spectral techniques - Definition and analysis of distances between sequences, structures, and images - Applications SUBMISSION PROCEDURE Only electronic submissions will be accepted. Manuscripts must be submitted via the special issue's web page http://www.dsi.unive.it/Similarity-PR/ The manuscripts must be submitted by December 15, 2004, and should conform to the standard guidelines of the Pattern Recognition journal. All submitted papers will be reviewed by at least three independent reviewers. -- Andrea Torsello PhD Dipartimento di Informatica, Universita' Ca' Foscari di Venezia via Torino 155, 30172 Venezia Mestre, Italy Tel: +39 0412348468 Fax: +39 0412348419 http://www.dsi.unive.it/~atorsell From lendasse at JAMES.HUT.FI Mon May 31 09:58:03 2004 From: lendasse at JAMES.HUT.FI (Amaury Lendasse) Date: Mon, 31 May 2004 16:58:03 +0300 Subject: Master Degree Thesis Position Message-ID: Dear Connectionists, I am looking for a Master Degree student. The position is described in http://www.cis.hut.fi/~lendasse/masterdegreeposition.html Dr. Amaury Lendasse *** PLEASE FORWARD TO POTENTIAL CANDIDATES *** MASTER DEGREE THESIS in the field of Time Series Prediction We propose a Master Degree Thesis in the field of Time series prediction and principally for the long-term prediction. These fields are briefly described below. This Master Degree Thesis should be done between September 2004 and December 2005. PROFILE: Applicants should have a Bachelor degree (or equivalent), in fields, such as Computer Science, Electrical Engineering, or Applied Math, or an equivalent or similar background. Knowledge about Neural Networks, Time Series Prediction, or System Identification is an advantage, but not a requirement. Ability to work in team is expected. The candidate will also have the following desirable qualifications: - strong programming skills (especially with Matlab); - good communication skills in English. APPLICATION FORM: Please send your application including a CV via E-mail to the contact person. CONTACT: Dr. Amaury Lendasse Helsinki University of Technology Laboratory of Computer and Information Science P.O. Box 5400 FIN-02015 HUT FINLAND tel. +358-9-451 4499 fax +358-9-451 3277 cell +358-40-770 0237 Email: lendasse at cis.hut.fi URL: http://www.cis.hut.fi/~lendasse Time series Prediction Time series forecasting is a challenge in many fields. In finance, one forecasts stock exchange indices or stock market indices; data processing specialists forecast the flow of information on their networks; producers of electricity forecast the load of the following day. The common point to their problems is the following: how can one analyze and use the past to predict the future? Many techniques exist: linear methods such as ARX, ARMA, etc., and nonlinear ones such as artificial neural networks. In general, these methods try to build a model of the process that is to be predicted. The model is then used on the last values of the series to predict future ones. The common difficulty to all methods is the determination of sufficient and necessary information for a good prediction. If the information is insufficient, the forecasting will be poor. On the contrary, if information is useless or redundant, modeling will be difficult or even skewed. Long-Term Prediction of Time Series As this problem can be found in many fields, many methods have been developed with very different approaches, from statistics to system identification and more recently neural networks. Most of the time, the models are linear and perform well on a rather short-term horizon, depending on the complexity of the problem. Their efficiency on a longer term is more questionable. This fact is due to the learning strategy used to fit the model to the data. The goal is usually to optimize the performance at a given term, most often the next time step. There are only a few attempts to explicitly predict values at long term, or at least global trends. This problem is quite hard since the uncertainty increases with the horizon of prediction. Another issue generally shared by classical models (such as ARX, ARMAX, .) is that they are used to predict a single value of a scalar time series. In practice some industrial applications require the prediction of a set of values in one single step instead of several independent values. Forecasting a vector of values requires more complex models able to predict several components together. If the approach is to develop several simple models and combine them to predict a vector, one can lose the correlation information between the vector components. Though each model may perform well, the forecasting accuracy could be rather poor when considering the vector of predicted values as a whole. Developing methods able to predict several values at each step, with the same expected performance on each value, should thus be a major concern. Let us consider now the general problem of forecasting at long term. Despite the fact that long-term predictions in real situations will probably never be very accurate, in some applications there is a need to have at least some ideas about the future of the time series. For example, answers to questions such as "Are there bounds on the future values?" or "What can we expect in average?" or even "Are the confidence intervals on future values large or narrow?" can give some ideas about the time series evolution at long-term. Publications about Times Predictions can be found in http://www.cis.hut.fi/~lendasse/Publications.html From nips04pub at hotmail.com Mon May 3 18:56:54 2004 From: nips04pub at hotmail.com (John Platt) Date: Mon, 3 May 2004 15:56:54 -0700 Subject: NIPS 2004 Call for Papers Message-ID: CALL FOR PAPERS --- NIPS 2004 Neural Information Processing Systems --- Natural and Synthetic Monday, December 13 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc Deadline for Paper Submissions: June 4, 2004 Submissions are solicited for the eighteenth annual meeting of an interdisciplinary NIPS Conference (December 14-16) which brings together researchers interested in all aspects of neural and statistical processing and computation. The Conference will include invited talks as well as Oral and Poster Presentations of refereed papers and Demonstrations. It is single track and highly selective. Preceding the main Conference will be one day of Tutorials (December 13), and following it will be two days of Workshops at Whistler/Blackcomb ski resort (December 17-18). INVITED SPEAKERS: * John Donoghue, Brown University --- Mind over Movement: Developing Neurotechnologies to Restore Lost Function; * Gerd Gigerenzer, Max Planck Institute for Human Development --- Fast and Frugal Heuristics: The Adaptive Toolbox; * Nati Linial, Hebrew University of Jerusalem --- Expanders, Eigenvalues and All That; * Bernard Palsson, University of California, San Diego, Systems Biology --- Bringing Genomes to Life; * Shimon Ullman, Weizmann Institute of Science --- Classification, Recognition and Segmentation Using Fragments Hierarchy TUTORIAL SPEAKERS: * William Bialek, Princeton University --- Optimization Principles in Neural Coding and Computation; * Shahar Mendelson, Australian National University --- A Geometric Approach to Statistical Learning Theory; * Radford Neal, University of Toronto --- Bayesian Methods in Machine Learning; * David Parkes, Harvard University --- Computational Mechanism Design and Auctions * Richard Szeliski, Microsoft Research --- Acquiring Detailed 3D Models from Images and Video; * Daniel Wolpert, University College London --- Probabilistic Computations in Human Sensoriomotor Control SUBMISSIONS Papers are solicited in all areas of neural information processing, including (but not limited to) the following: * Algorithms and Architectures --- statistical learning algorithms, neural networks, kernel methods, graphical models, Gaussian processes, independent component analysis, model selection, combinatorial optimization. * Applications --- innovative applications or fielded systems that use machine learning, including systems for time series prediction, bioinformatics, text/web analysis, multimedia processing, and robotics. * Brain Imaging --- neuroimaging, cognitive neuroscience, EEG (electroencephalogram), ERP (event related potentials), MEG (magnetoencephalogram), fMRI (functional magnetic resonance imaging), brain mapping, brain segmentation. * Cognitive Science and Artificial Intelligence --- theoretical, computational, or experimental studies of perception, psychophysics, human or animal learning, memory, reasoning, problem solving, language, and neuropsychology. * Control and Reinforcement Learning --- decision and control, exploration, planning, navigation, Markov decision processes, game-playing, multi-agent coordination, computational models of classical and operant conditioning. * Emerging Technologies --- analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, microrobotics, bioMEMS, neural prostheses, photonics, molecular and quantum computing. * Learning Theory --- generalization and regularization, information theory, statistical physics of learning, Bayesian methods, approximation bounds, online learning and dynamics. * Neuroscience --- theoretical and experimental studies of processing and transmission of information in biological neurons and networks, including spike train generation, synaptic modulation, plasticity and adaptation. * Speech and Signal Processing --- recognition, coding, synthesis, denoising, segmentation, source separation, auditory perception, psychoacoustics, dynamical systems, recurrent networks, Markov models. * Visual Processing --- biological and machine vision, image processing and coding, segmentation, object detection and recognition, motion detection and tracking, visual psychophysics, visual scene analysis and interpretation. * Demonstrations --- Authors wishing to submit to the Demonstration track should consult the Conference web site. REVIEW CRITERIA: Submissions will be refereed on the basis of technical quality, novelty, significance, and clarity. Authors new to NIPS are particularly encouraged to submit. There will be an opportunity after the meeting to revise accepted manuscripts. PAPER FORMAT: Submissions may be up to eight pages in length, including figures and references, using a font no smaller than 10 point. Text is to be confined within a 8.25 inch by 5 inch rectangle. Submissions violating these guidelines will not be considered. SUBMISSION INSTRUCTIONS: NIPS accepts only electronic submissions in postscript and PDF format. The Conference web site will accept electronic submissions from May 19, 2004 until midnight, June 4, 2004, Pacific daylight time. ORGANIZING COMMITTEE: * General Chair, Lawrence Saul, University of Pennsylvania; * Program Chair, Yair Weiss, Hebrew University of Jerusalem; * Tutorials Chair, Richard Zemel, University of Toronto; * Workshops Co-chairs, Satinder Singh Baveja, University of Michigan, Daniel Lee, University of Pennsylvania; * Demonstrations Chairs, Tobi Delbruck, ETH/University of Zurich, Timothy Horiuchi, University of Maryland; * Publications Chair, Leon Bottou, NEC Research Institute; * Publicity Chair, John Platt, Microsoft Research; * Online Proceedings Chair, Andrew McCallum, University of Massachusetts Amherst; * Volunteers Chair, Sam Roweis, University of Toronto. PROGRAM COMMITTEE: Yair Weiss (Chair), Hebrew University of Jerusalem; Yoshua Bengio, University of Montreal; Jeff Bilmes, University of Washington; Nello Cristianini, University of California Davis; Trevor Darrell, Massachusetts Institute of Technology; Geoff Gordon, Carnegie Mellon University; Daphne Koller, Stanford University; John Lafferty, Carnegie Mellon University; Shih-Chii Liu, Swiss Federal Institute of Technology Zurich; Ron Meir, Technion, Israeli Institute of Technology; Maneesh Sahani, University College London; Bernhard Schoelkopf, Max-Planck Institute for Biological Cybernetics; Matthias Seeger, University of California, Berkeley; Richard Shiffrin, Indiana University; Eero Simoncelli, New York University; Martin Wainwright, University of California Berkeley; Daphna Weinshall, Hebrew University of Jerusalem. DEADLINE FOR PAPER SUBMISSIONS: JUNE 4, 2004 From jebara at cs.columbia.edu Tue May 4 13:23:12 2004 From: jebara at cs.columbia.edu (Tony Jebara) Date: Tue, 4 May 2004 13:23:12 -0400 Subject: ICDL 2004 - Extension to May 21, 2004 Message-ID: ICDL 2004 CALL FOR PAPERS DEADLINE HAS BEEN EXTENDED TO FRIDAY, MAY 21 2004 There will be absolutely no further extensions. THIRD INTERNATIONAL CONFERENCE ON DEVELOPMENT AND LEARNING: DEVELOPING SOCIAL BRAINS The Salk Institute October 20-22, 2004 San Diego, California A Satellite Conference preceding The Annual Society for Neuroscience Conference http://www.icdl.cc The goal of the conference is to bring together leading researchers in neuroscience, machine learning, robotics, and developmental psychology, in order to gain new insights about learning and development in natural organisms and robots. The scope of developmental processes to be considered is broad, including cognitive, social, emotional, and many other skills exhibited by humans, and animals. The theme of the conference this year will be "Developing Social Brains", but other topics related to development and learning are welcome. PAPER SUBMISSION The extended submission deadline is May 21, 2004. Papers for the meeting can be submitted ONLY through the conference's web site at: http://www.icdl.cc. Papers can be submitted either as a 200 word summary or as a full paper (max 8 typeset pages). IMPORTANT: There will be NO further submission deadline extensions. SPECIAL ISSUE ON NEUROCOMPUTING Authors of selected papers will be invited to submit an extended version of their paper for publication in a special issue of the Neurocomputing Journal, on Development, Learning, and the Social Brain, published by Elsevier Science B.V. (http://www.elsevier.nl/locate/neucom) INVITED TALKS (Not yet confirmed) John Allman Dana Ballard Rodney Brooks Eric Courchesne Peter Dayan Jeff Elman William Greenough James L. McClelland Pietro Perona Terrence Sejnowski Joan Stiles John Watson REVIEW PROCESS All submitted papers will be reviewed by the program committee. Papers will be judged and accepted for the meeting based on the clarity with which the work is described and the relevance to the goals of the conference. All accepted papers not selected for oral talks as well as papers explicitly submitted as poster presentations will be included in one of three evening poster sessions. Authors will be notified of the presentation format of their papers by the beginning of August. ORGANIZING COMMITTEE: General Chair: Javier R. Movellan: Co-Chairs: Andrea Chiba, Gedeon Deak, Jochen Triesch. Program Chair: Jochen Triesch and Tony Jebara. Program Co-Chairs: Marian Stewart-Bartlett, Gwen Ford Littlewort. Publications Chair: Gedeon Deak. ADVISORY BOARD: Jeff Elman Pat Langley James L. McClelland Sandy Pentland Terrence Sejnowski Mriganka Sur Esther Thelen Juyang Weng PROGRAM COMMITTEE: Minoru Asada Dana Ballard Luis Baumela Simon Baron-Cohen Mark Baxter Jeff Cohn Kerstin Dautenhahn Kenji Doya Martha Farah Teresa Farroni Masahiro Fujita Ann Graybiel William Greenough Michael Hasselmo Shoji Itakura Hiroshi Ishiguro Robert Jacobs David Kleinfeld Mark Konishi Denis Mareshal Risto Miikulainen Douglas Nitz Roz Picard Steven Quartz Rajesh Rao Matthew Schlesinger Gregor Schoener Geoffrey Schoenbaum Linda Smith Olaf Sporns Luc Steels Valerie Stone Manuela Veloso Paul Verschure Christoph von der Malsburg Hiroyuki Yano CONFERENCE REGISTRATION: Register online at http://www.icdl.cc Student Registration is: $150 Non-student Registration is: $290 From cindy at bu.edu Wed May 5 10:13:11 2004 From: cindy at bu.edu (Cynthia Bradford) Date: Wed, 5 May 2004 10:13:11 -0400 Subject: 8th ICCNS: Final Call for Registration Message-ID: <00a801c432ab$18f19b20$903dc580@cnspc31> Apologies if you receive more than one copy of this email. ***** FINAL CALL FOR REGISTRATION ***** EIGHTH INTERNATIONAL CONFERENCE ON COGNITIVE AND NEURAL SYSTEMS May 19-22, 2004 http://cns.bu.edu/cns-meeting/conference.html Boston University CNS Department 677 Beacon Street Boston, Massachusetts 02215 Sponsored by Boston University's Center for Adaptive Systems and Department of Cognitive and Neural Systems with financial support from the Office of Naval Research This single-track conference, which offers invited lectures and contributed lectures and posters, focuses on two main themes: How does the brain control behavior? How can technology emulate biological intelligence? CONFIRMED INVITED SPEAKERS: Tutorial Lecture Series: Stephen Grossberg Keynote Lectures: John Anderson and Miguel Nicolelis Invited Speakers: Ehud Ahissar, Alan D. Baddeley, Moshe Bar, Gail A. Carpenter, Stephen Goldinger, Daniel Kersten, Stephen M. Kosslyn, Tai-Sing Lee, Eve Marder, Bartlett W. Mel, Jeffrey D. Schall, Chantal Stern, Mriganka Sur, Joseph Z. Tsien, William H. Warren Jr., Jeremy Wolfe. Please visit the web site for conference details, including: --the registration form http://cns.bu.edu/cns-meeting/registration.html --a schedule of the oral and poster presentations http://cns.bu.edu/cns-meeting/schedule.html --details about the tutorial lecture series http://cns.bu.edu/cns-meeting/grossberg-lectures.html --local lodging options http://cns.bu.edu/cns-meeting/hotels.html From gregoire.dubois at jrc.it Thu May 6 03:19:38 2004 From: gregoire.dubois at jrc.it (Gregoire Dubois) Date: Thu, 6 May 2004 09:19:38 +0200 Subject: Spatial Interpolation Comparison exercise: who is going to do the "best map"? Message-ID: <000001c4333a$7daf0b80$9badbf8b@ei.jrc.it> Please, find hereafter the announcement for the following international exercise. I would very much appreciate the contribution of the community of connectionists. SIC2004 is the second edition of a scientific exercise dedicated to decision support systems and spatial statistics. SIC2004 stands for Spatial Interpolation Comparison 2004. Participants will receive a subset of an environmental data set (typically measurements of an environmental variable + spatial coordinates of the sampling places) and will have to estimate the values taken by the variable at the remaining locations of the full data set. The true values found at these locations will be made public only at the end of the exercise. Various criteria will be used to assess the performances of the interpolation algorithms (time of calculation, minimum errors, etc.). This edition will focus on automatic mapping algorithms: participants to SIC2004 will have to prepare their algorithms before receiving the data (only sampling locations will be given) and no interaction with the algorithm will be allowed during the exercise. Participants to SIC2004 will be invited to submit a manuscript at the end of the exercise for publication in the online journal GIDA (Geographic Information and Decision Analysis) as well as in a European Report (hardcopy). For more information, please visit the web site www.ai-geostats.org Thank you very much for your attention, Gregoire Keywords: geostatistics,spatial statistics, neural networks, spatial interpolation, GIS. PS: in the case you would like to support such an exercise, could you please forward this information to other potential scientific communities (e.g. machine learning, neural networks, statisticians, etc.). Thank you very much. __________________________________________ Gregoire Dubois (Ph.D.) JRC - European Commission Radioactivity Environmental Monitoring TP 441, Via Fermi 1 21020 Ispra (VA) ITALY Tel. +39 (0)332 78 6360 Fax. +39 (0)332 78 5466 Email: gregoire.dubois at jrc.it WWW: http://www.ai-geostats.org WWW: http://rem.jrc.cec.eu.int From wsenn at cns.unibe.ch Fri May 7 09:06:32 2004 From: wsenn at cns.unibe.ch (Walter Senn) Date: Fri, 07 May 2004 15:06:32 +0200 Subject: Postdoc and PhD position available in computational neuroscience Message-ID: <409B89D8.5010007@cns.unibe.ch> A postdoc and a PhD position is available by October 1, 2004, in the computational neuroscience group at the Institute of Physiology, University of Bern, Switzerland. We are studying biophysical models of neurons and synapses in the context of cortical information processing and learning. The aim of our reaearch is to relate dynamic properties of isolated neurons and synapses (studied in in vitro experiments by the groups of H.-R. Luescher and M.E. Larkum in our Institute) to higher cognitive functions. In our recent work we described slow adaptation of cortical neurons as measured in vitro in our lab [1], and related it to the cortical representation of time [2]. Other experimental and theoretical research in our lab shows that top-down input onto dendrites of layer 5 pyramidal cells can modulate the gain of these neurons [3], as observed in attentional experiments in vivo. In a current project we are studying spike-timing dependent plasticity (STDP) in the context of pattern classification and reinforcement learning [4]. These algorithms will be tested against data from in vitro experiments on long-term plasticity performed in our lab. Ideal candidates have a strong background in neural networks and/or computational neuroscience, and are interested to work at the interface between computational theories and electrophysiology. Please send CV, addresses of three references, and a half page description of your work and interests to wsenn at cns.unibe.ch (but no doc-files, if possible). Walter Senn and Stefano Fusi http://www.cns.unibe.ch/ [1] Neocortical pyramidal cells respond as integrate-and-fire neurons to in vivo-like input currents. A. Rauch, G. La Camera, H-R L?scher, W. Senn and S. Fusi. Journal of Neurophysiology, 90:1598-1612 (2003) http://www.cns.unibe.ch/publications/ftp/paper_transferfunction.pdf [2] Climbing neuronal activity as an event-based cortical representation of time. J. Reutimann, V. Yakovlev, S. Fusi and W. Senn. The Journal of Neuroscience, 24:3295-3303 (2004) http://www.cns.unibe.ch/publications/ftp/paper_climbing_activity.pdf [3] Top-down dendritic input increases the gain of layer 5 pyramidal neurons. M. Larkum, W. Senn and H.-R. L?scher, Cerebral Cortex, in press (2004) http://www.cns.unibe.ch/publications/ftp/paper_gain_modulation04.pdf [4] Learning with bounded synapses generates synaptic democracy and balanced neurons. With a proof of the perceptron convergence theorem for bounded synapses. W. Senn and S. Fusi, submitted (2004) http://www.cns.unibe.ch/publications/ftp/paper_bounded_synapses04.pdf -- Walter Senn Phone office: +41 31 631 87 21 Physiological Institute Phone home: +41 31 332 38 31 University of Bern Fax: +41 31 631 46 11 Buehlplatz 5 Email: wsenn at cns.unibe.ch CH-3012 Bern http://www.cns.unibe.ch/~wsenn/ From dejong at cs.uu.nl Tue May 11 08:06:33 2004 From: dejong at cs.uu.nl (Edwin de Jong) Date: Tue, 11 May 2004 14:06:33 +0200 (MEST) Subject: Preprint available: Ideal Evaluation from Coevolution Message-ID: Dear colleagues, I would like to announce a preprint of the following article: De Jong, E.D. and Pollack, J.B. (2004). Ideal Evaluation from Coevolution. Evolutionary Computation, Vol. 12, no. 2 (to appear). Available from: http://www.cs.uu.nl/~dejong/#coevec Abstract: In many problems of interest, performance can be evaluated using tests, such as examples in concept learning, test points in function approximation, and opponents in game-playing. Evaluation on all tests is often infeasible. Identification of an accurate evaluation or fitness function is a difficult problem in itself, and approximations are likely to introduce human biases into the search process. Coevolution {\em evolves} the set of tests used for evaluation, but has so far often led to inaccurate evaluation. We show that for any set of learners, a Complete Evaluation Set can be determined that provides ideal evaluation as specified by Evolutionary Multi-Objective Optimization. This provides a principled approach to evaluation in coevolution, and thereby brings {\em automatic} ideal evaluation within reach. The Complete Evaluation Set is of manageable size, and progress towards it can be accurately measured. Based on this observation, an algorithm named DELPHI is developed. The algorithm is tested on problems likely to permit progress on only a subset of the underlying objectives. Where all comparison methods result in overspecialization, the proposed method and a variant achieve sustained progress in all underlying objectives. These findings demonstrate that ideal evaluation may be approximated by practical algorithms, and that accurate evaluation for test-based problems is possible even when the underlying objectives of a problem are unknown. Any comments you may have on this work are very welcome. Best regards, Edwin de Jong -------------------------------------------------------------------- http://www.cs.uu.nl/~dejong From poznan at iub-psych.psych.indiana.edu Tue May 11 13:59:00 2004 From: poznan at iub-psych.psych.indiana.edu (Roman Poznanski) Date: Tue, 11 May 2004 12:59:00 -0500 Subject: Special Issue of JIN Dedicated to the Late Gen Matsumoto Message-ID: <40A11464.9000908@iub-psych.psych.indiana.edu> JOURNAL OF INTEGRATIVE NEUROSCIENCE VOLUME 3, ISSUE 2 (JUNE, 2004) Editorial: M. Ichikawa Prologue- the work of the late Professor Gen Matsumoto: K.Aihara, I.Tsuda, and H.Fujii Short Communications 1. I. TASAKI ON THE CONDUCTION VELOCITY OF NERVE FIBERS 2. M. ICHIKAWA and G. MATSUMOTO THE BRAIN-COMPUTER: ORIGIN OF THE IDEA AND PROGRESS IN ITS REALIZATION 3. H. TSUJINO OUTPUT-DRIVEN OPERATION AND MEMORY-BASED ARCHITECTURE PRINCIPLES EMBEDDED IN A REAL WORLD DEVICE. Research Reports 1. Y. YAMAGUCHI, Y.AOTA, N.SATO, H.WAGATSUMA, and Z.WU SYNCHRONIZATION OF NEURAL OSCILLATIONS AS A POSSIBLE MECHANISM UNDERLYING EPISODIC MEMORY: A STUDY OF THETA RHYTHM IN THE HIPPOCAMPUS 2. I. TSUDA, H.FUJII, S.TADOKORO, T. YASUOKA, and Y.YAMAGUTI CHAOTIC ITINERANCY AS A MECHANISM OF IRREGULAR CHANGES BETWEEN SYNCHRONIZATION AND DESYCHCHRONIZATION IN A NEURAL NETWORK. 3. H.FUJII, K. AIHARA, and I.TSUDA FUNCTIONAL RELEVANCE OF 'EXCITATORY' GABA ACTIONS IN CORTICAL INTERNEURONS: A DYNAMICAL SYSTEMS APPROACH. 4. S.DOI, J.INOUE, and S. KUMAGAI CHAOTIC SPIKING IN THE HODGKIN-HUXLEY NERVE MODEL WITH SLOW INACTIVATION OF THE SODIUM CURRENT 5. R.R. POZNANSKI ANALYTICAL SOLUTIONS OF THE FRANKENHAEUSER-HUXLEY EQUATIONS 6. P. BASSER SCALING LAWS FOR MYELINATED AXONS DERIVED FROM AN ELECTROTONIC CORE-CONDUCTOR MODEL ------------------------------------------------------------------------ For orders within Europe, please contact the Imperial College Press sales department at: Tel: +44 (0)20 7836-0888 Fax: +44 (0)20 7836-2020 during U.K. business hours. Outside Europe, our books and journals are distributed by World Scientific Publishing Co. World Scientific Publishing Co. Pte. Ltd. 5 Toh Tuck Link, SINGAPORE 596224 Fax: 65-6467-7667 Tel: 65-6466-5775 E-mail: wspc at wspc.com.sg Price Information: ISSN: 0219-6352 ; Vol. 3/2004; 4 Issues Special Rates: Individuals -- Roman R. Poznanski, Ph.D Associate Editor, Journal of Integrative Neuroscience Department of Psychology Indiana University 1101 E. 10th St. Bloomington, IN 47405-7007 email: poznan at iub-psych.psych.indiana.edu phone (Office): (812) 856-7195 http://www.worldscinet.com/jin/mkt/editorial.shtml From masulli at disi.unige.it Tue May 11 04:57:08 2004 From: masulli at disi.unige.it (Francesco Masulli) Date: Tue, 11 May 2004 10:57:08 +0200 Subject: CFP: CIBB 2004 - INTERNATIONAL MEETING ON COMPUTATIONAL INTELLIGENCE METHODS FOR BIOINFORMATICS AND BIOSTATISTICS Message-ID: <200405111057.08616.masulli@disi.unige.it> CIBB 2004 INTERNATIONAL MEETING ON COMPUTATIONAL INTELLIGENCE METHODS FOR BIOINFORMATICS AND BIOSTATISTICS September 14-15, 2004, Perugia, ITALY CIBB 2004 runs as a special session of WIRN 04 - XV ITALIAN WORKSHOP ON NEURAL NETWORKS http://siren.dsi.unimi.it/conferences/SI/html/index.html FIRST CALL FOR PAPERS JOINTLY ORGANIZED BY - BIOPATTERNS - European Network of Excellence on Computational Intelligence for Biopattern analysis in Support of eHealthcare - SIREN - Italian Neural Networks Society AIMS CIBB 2004 addresses a cutting edge area of application of Neural Networks, Fuzzy Logic and Evolutionary Computation methods. Technical areas include, but are not limited to: - Data and methods for prognosis - Data and methods for diagnosis - Integration of clinical and genetic data - Proteomics - Pharmacogenetics CHAIRS Francesco Masulli, University of Pisa (Italy) Antonina Starita, University of Pisa (Italy) Roberto Tagliaferri, University of Salerno (Italy) IMPORTANT DATES Paper electronic submission deadline: June 15 2004 Notification of acceptance: July 15 2004 Submission of camera-ready papers: Sept 2 2004 SUBMISSIONS Papers must not be longer than 6 pages, including a cover sheet stating (1) Paper title; (2) Keywords; (3) Authors names and affiliations; (4) Corresponding author's name and contact details, including telephone/fax numbers and e-mail address. For electronic submission refer to http://siren.dsi.unimi.it/conferences/SI/html/index.html All accepted papers submitted by registered participants to WIRN 2004 will be included in the proceedings book of WIRN 2004, that will be published by an international Publisher. REGISTRATION FEES The registration to WIRN 2004 includes the free participation at CIBB 2004. ------------------------------------------------------- From terry at salk.edu Wed May 12 14:29:23 2004 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 12 May 2004 11:29:23 -0700 (PDT) Subject: NEURAL COMPUTATION 16:6 In-Reply-To: <200404140447.i3E4leM16495@dax.salk.edu> Message-ID: <200405121829.i4CITNt96552@dax.snl.salk.edu> Neural Computation - Contents - Volume 16, Number 6 - June 1, 2004 LETTERS Nonlinear Population Codes Maoz Shamir and Haim Sompolinsky Enhancement of Information Transmission Efficiency by Synaptic Failures Mark S. Goldman An Algorithm for the Detection of Faces on the Basis of Gabor Features and Information Maximization Hitoshi Imaoka and Kenji Okajima Analysis of Sparse Representation and Blind Source Separation Yuanqing Li, Andrzej Cichocki and Shun-ichi Amari Minimax Mutual Information Approach for Independent Component Analysis Deniz Erdogmus, Kenneth E. Hild II, Yadunandana N. Rao and Jose C. Principe Improving Generalization Capabilities of Dynamic Neural Networks Miroslaw Galicki, Lutz Leistritz , Ernst Bernhard Zwick and Herbert Witte A Modified Algorithm for Generalized Discriminant Analysis Wenming Zheng, Li Zhao and Ciarong Zou Stability-Based Validation of Clustering Solutions Tilman Lange, Volker Roth, Mikio L. Braun and Joachim M. Buhmann ----- 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 dimi at ci.tuwien.ac.at Wed May 12 07:51:32 2004 From: dimi at ci.tuwien.ac.at (Evgenia Dimitriadou) Date: Wed, 12 May 2004 13:51:32 +0200 (CEST) Subject: CI BibTeX Collection -- Update Message-ID: The following volumes have been added to the collection of BibTeX files maintained by the Vienna Center for Computational Intelligence: IEEE Transactions on Evolutionary Computation, Volumes 7/5 - 8/2 IEEE Transactions on Fuzzy Systems, Volumes 11/5 - 12/2 IEEE Transactions on Neural Networks, Volumes 14/5 - 15/3 Machine Learning, Volumes 53/3 - 55/2 Neural Computation, Volumes 15/10 - 16/5 Neural Networks, Volumes 16/7 - 17/4 Neural Processing Letters, Volumes 17/3 - 19/2 Journal of Statistical Software Volumes 1/1 - 10/5 Most files have been converted automatically from various source formats, please report any bugs you find. The complete collection can be downloaded from http://www.ci.tuwien.ac.at/services/BibTeX.html ftp://ftp.ci.tuwien.ac.at/pub/texmf/bibtex/ Best, Vivi ************************************************************************ * Evgenia Dimitriadou * ************************************************************************ * Institut fuer Statistik * Tel: (+43 1) 58801 10773 * * Technische Universitaet Wien * Fax: (+43 1) 58801 10798 * * Wiedner Hauptstr. 8-10/1071 * Evgenia.Dimitriadou at ci.tuwien.ac.at * * A-1040 Wien, Austria * http://www.ci.tuwien.ac.at/~dimi* ************************************************************************ From qian at brahms.cpmc.columbia.edu Thu May 13 18:53:22 2004 From: qian at brahms.cpmc.columbia.edu (Ning Qian) Date: Thu, 13 May 2004 18:53:22 -0400 Subject: papers on stereo and motor modeling available Message-ID: <40A3FC62.8090701@brahms.cpmc.columbia.edu> Dear Colleagues, The pdf files of two preprints from my lab are available at the links below. The first paper is about a multi-scale and multi-orientation stereo model, and the second is on comparing models of motor planning. The pdf files of some old papers, including the one on protein structure prediction that several people asked me for, can also be find at the same site http://brahms.cpmc.columbia.edu . Best regards, Ning ----------------------------------------------------------------- A Coarse-to-fine Disparity Energy Model with both Phase-shift and Position-shift Receptive Field Mechanisms Yuzhi Chen and Ning Qian, Neural Computation, 2004, in press. Numerous studies suggest that the visual system uses both phase- and position-shift receptive field (RF) mechanisms for the processing of binocular disparity. Although the difference between these two mechanisms has been analyzed before, previous work mainly focused on disparity tuning curves instead of population responses. However, tuning curve and population response can exhibit different characteristics, and it is the latter that determines disparity estimation. Here we demonstrate, in the framework of the disparity energy model, that for relatively small disparities, the population response generated by the phase-shift mechanism is more reliable than that generated by the position-shift mechanism. This is true over a wide range of parameters including the RF orientation. Since the phase model has its own drawbacks of underestimating large stimulus disparity and covering only a restricted range of disparity at a given scale, we propose a coarse-to-fine algorithm for disparity computation with a hybrid of phase-shift and position-shift components. In this algorithm, disparity at each scale is always estimated by the phase-shift mechanism to take advantage of its higher reliability. Since the phase based estimation is most accurate at the smallest scale when the disparity is correspondingly small, the algorithm iteratively reduces the input disparity from coarse to fine scales by introducing a {\em constant} position-shift component to all cells for a given location in order to offset the stimulus disparity at that location. The model also incorporates orientation pooling and spatial pooling to further enhance reliability. We have tested the algorithm on both synthetic and natural stereo images, and found that it often performs better than a simple scale averaging procedure. http://brahms.cpmc.columbia.edu/publications/stereo-scale.pdf Different Predictions by the Minimum Variance and Minimum Torque-Change Models on the Skewness of Movement Velocity Profiles Hirokazu Tanaka, Meihua Tai and Ning Qian, Neural Computation, 2004, in press. We investigated the differences between two well-known optimization principles for understanding movement planning: the minimum variance (MV) model of Harris \& Wolpert and the minimum torque-change (MTC) model of Uno et al. Both models accurately describe the properties of human reaching movements in ordinary situations (e.g. nearly straight paths and bell-shaped velocity profiles). However, we found that the two models can make very different predictions when external forces are applied or when the movement duration is increased. We considered a second-order linear system for the motor plant that has been used previously to simulate eye movements and single-joint arm movements, and were able to derive analytical solutions based on the MV and MTC assumptions. With the linear plant, the MTC model predicts that the movement velocity profile should always be symmetric, independent of the external forces and movement duration. In contrast, the MV model strongly depends on the movement duration and the system's degree of stability; the latter in turn depends on the total forces. The MV model thus predicts a skewed velocity profile under many circumstances. For example, it predicts that the peak location should be skewed toward the end of the movement when the movement duration is increased in the absence of any elastic force. It also predicts that with appropriate viscous and elastic forces applied to increase the system stability, the velocity profile should be skewed toward the beginning of the movement. The velocity profiles predicted by the MV model can even show oscillations when the plant becomes highly oscillatory. Our analytical and simulation results suggest specific experiments for testing the validity of the two models. http://brahms.cpmc.columbia.edu/publications/motor-models.pdf -- Ning Qian, Ph. D. Associate Professor Ctr. Neurobiology & Behavior Columbia University / NYSPI Kolb Annex, Rm 519 1051 Riverside Drive, Box 87 New York, NY 10032, USA http://brahms.cpmc.columbia.edu nq6 at columbia.edu 212-543-6931 ext 600 (Office) 212-543-5816 (Fax) From l.wiskott at biologie.hu-berlin.de Thu May 13 06:09:14 2004 From: l.wiskott at biologie.hu-berlin.de (Laurenz Wiskott) Date: Thu, 13 May 2004 12:09:14 +0200 Subject: PhD position available in Berlin Message-ID: <16547.18762.414079.62270@huxley.biologie.hu-berlin.de> Please forward this job advertisement to students who might be interested. Thanks, Laurenz Wiskott. ___________________________________________________________________________ Open Position for a PhD-Student in Computational Neuroscience at the Institute for Theoretical Biology, Berlin http://itb.biologie.hu-berlin.de/~wiskott/jobs.html ___________________________________________________________________________ Institute: Institute for Theoretical Biology Humboldt-Universit?t zu Berlin Invalidenstra?e 43 D-10115 Berlin, Germany http://itb.biologie.hu-berlin.de/ The Institute for Theoretical Biology is a young and dynamic lab with three full professors, three junior research groups, and about 60 students and researchers doing interdisciplinary and innovative research in different areas of theoretical biology. Research group: The position is available in the junior research group led by Dr. Laurenz Wiskott and funded by the Volkswagen Foundation. Research topics: The central issue addressed by the junior research group is the question of how invariant object perception can be achieved and learned in sensory systems, especially in the primate visual system. The project for this PhD position will be a theoretical analysis of simulation results. Teaching: There will be some duties as a teaching assistant, e.g. for the basic mathematics courses. Time: The position is available now. The initial appointment will be for 1 year with a possible extension by additional 2 years. Requirements: Candidates should have a recent degree (Diploma/MSc) in mathematics or theoretical physics. Required are strong mathematical and programming skills as well as the ability to communicate and work well in a team. Salary: Salary will be ? BAT IIa-O and will depend on age, and family status. BAT is the regular salary scale for public employees in Germany. Inquiries: Informal inquiries can be addressed to Dr. Laurenz Wiskott. Application: Complete applications should be sent to Dr. Laurenz Wiskott at the address given above. Please send only copies and not original documents, since the applications will not be sent back. You can also send applications via email, but please make sure they are complete and in a convenient format. Handicapped applicants with corresponding qualifications will be considered preferentially. To increase the proportion of female scientists, applications of qualified females are especially welcome. Deadline: None. Applications will be accepted until the position is filled, as will be indicated on the web-page of this job advertisement; see below. WWW: The following web page contains additional information: http://itb.biologie.hu-berlin.de/~wiskott/jobs.html. From C.Campbell at bristol.ac.uk Thu May 13 07:33:48 2004 From: C.Campbell at bristol.ac.uk (ICG Campbell, Engineering Mathematics) Date: Thu, 13 May 2004 12:33:48 +0100 Subject: RA position (Bristol University, UK) Message-ID: <562789.1084451628@ems-iggs.enm.bris.ac.uk> Research Assistant Position: Analysis of Microarray Data Applications are invited from experienced postdoctoral Research Fellows to work on a project principally focussed on new methods for the analysis and interpretation of microarray datasets. This project will involve the design of new algorithms and data analysis techniques to improve inference in medical diagnosis, predicted prognosis and fusion of data from different sources. We will use a variety of methods including Bayesian techniques and graphical methods, kernel methods and other approaches. This project is part of a large collaborative European Framework 6 consortium and will involve interaction and visits to collaborating groups in the UK and visits to overseas groups in Norway, France, Israel and the US. On the experimental side our collaborators are medical groups generating data for cancer research and diseases of the brain and the Research Fellow should be willing to sometimes assist in the interpretation and analysis of datasets generated by our collaborators. The Research Fellow will be hosted by the Computational Intelligence subgroup of the Artificial Intelligence Group within the Department of Engineering Mathematics, Bristol University. Details about our group and publications can be obtained from our website http://www.enm.bris.ac.uk/ai/ Applicants should have excellent mathematical and computational skills, and preferably a background in machine learning, statistics or bioinformatics. Details: Duration: 2 years with possible extension. Staff category and grade: RA1A Salary range: ?18893 - ?28279 per annum depending on qualifications and experience. Application: Informal enquiries should be made to Dr. Colin Campbell, email: C.Campbell(at)bris.ac.uk Formal applications and details of how to apply are available from the Personnel Services website, at http://www.bris.ac.uk/boris/jobs/ (vacancy ref. 10256). The closing date for applications is 30th June 2004. ---------------------- ICG Campbell, Engineering Mathematics C.Campbell at bristol.ac.uk From stefan.wermter at sunderland.ac.uk Fri May 14 06:05:57 2004 From: stefan.wermter at sunderland.ac.uk (Stefan Wermter) Date: Fri, 14 May 2004 11:05:57 +0100 Subject: NeuroBotics - Bioinspired computation for Robotics - Call for papers Message-ID: <40A49A05.1050305@sunderland.ac.uk> *************************************** International Workshop on NeuroBotics: Bioinspired Computation for Robotics **************************************** 20 September 2004 Call for Papers --------------- Substantial progress has been made recently in bio-inspired computation and robotics. This international workshop invites contributions to robotics which use methods of learning or artificial neural networks and/or are inspired by observations and results in neuroscience, cognitive science and animal behaviour. Topics of interest include but are not restricted to: ------- Neural networks for robots Biomimetic robots Learning for robotics Cognitive robots Speech interfaces and neural networks Neural Vision Talking robots Spiking neural networks in robots Learning self localization and mapping Imitation and neural networks Cognitive Development in robots etc Location -------- The workshop is organised as part of the 27th Conference on Artificial Intelligence (KI-2004) and runs parallel to the 34th Annual Meeting of the German Computer Science Society (Informatik 2004). The workshop will take place on September 20, 2004 at the University of Ulm, Germany. Format ------ We invite contributions for half-hour talks and, if possible, additional short demonstrations. If we receive sufficient submissions of high quality, we plan to publish the revised articles of the workshop contributions in a journal or book. Deadlines --------- Please send either abstracts of one page or full papers up to 8 pages (and if appropriate, possible descriptions of demonstrations) until June 9, 2004 to the organisers email address below. You will receive notification of acceptance, a more detailed workshop program and regarding the planned publication by July 5, 2004. Workshop registration: until August 20, 2004. Conference fees: will be not more than 100 Euro for the workshop. Registration at the conference will be optional. Organisation and more details: ------------------------------ More details about the AI-conference can be found at: http://ki2004.uni-ulm.de and more details about the organsing MirrorBot project to found at http://www.his.sunderland.ac.uk/mirrorbot/ and updates on this call at http://www.his.sunderland.ac.uk/mirrorbot/call.html Organisers ---------- Prof. Gnther Palm Neural Information Processing Computer Science Faculty University of Ulm Oberer Eselsberg D-89069 Ulm Germany palm at neuro.informatik.uni-ulm.de Prof. Stefan Wermter Hybrid Intelligent Systems School of Computing and Technology University of Sunderland St Peter's Way Sunderland SR6 0DD United Kingdom Stefan.Wermter at sunderland.ac.uk http://www.his.sunderland.ac.uk/~cs0stw/ From codrina.lauth at ais.fraunhofer.de Thu May 13 06:25:11 2004 From: codrina.lauth at ais.fraunhofer.de (Codrina Lauth) Date: Thu, 13 May 2004 12:25:11 +0200 Subject: Call for Demo Sessionat PKDD/ECML 2004: DEADLINE: 15 May 2004 Message-ID: <003d01c438d4$b035baf0$b4921a81@ais.fhg.de> Call for Demo Session The 15th European Conference on Machine Learning (ECML) and the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases (PKDD) will be co-located in Pisa, Italy, September 20-24, 2004. The combined event will comprise presentations of contributed papers and invited speakers, a wide program of workshops, tutorials, and a discovery challenge, and last but not least a demo session. The Demo Session is reserved to innovative applications or successful experiences/prototypes in applying the Knowledge Discovery Process to practical cases. Also, are solicited demonstrations of prototypes/systems which show the integration of new Data Mining and Machine Learning technologies, advances using complex data structures and databases, XML-data, multimedia, expert systems in innovative application domains, such as bioinformatics, medical domains, WEB, e-commerce, etc. Demonstration proposals should give a short description of the demonstrated system, explain what is going to be demonstrated, and state the significance of the contribution to the field of Machine Learning and Knowledge Discovery in Databases. Important for the participants, in addition to the description of technical functionalities, will be to provide also performance results and quality measures (on standard benchmarks in the field and comparative examples). Short papers describing demos will appear in the Proceedings. Demonstration proposals must be no more than three (3) pages, in English and should be formatted according to the Springer-Verlag Lecture Notes in Computer Science guidelines. Authors instructions and style files can be downloaded at http://www.springer.de/comp/lncs/authors.html. Important Dates Submission Proposal deadline: Saturday May 15, 2004 Notification of acceptance: Monday June 7, 2004 Camera-ready copies due: Monday June 28, 2004 (Strict deadline) In the exhibit space, exhibitors give demonstrations and present their prototypes/systems to the audience, answering to their questions. A selected demo is shown to the interested audience in a slot of two hours including a coffee break. Each demo is then repeated more times (till three) during all week. Demos are explicitely mentioned in the conference program scheduling and in the daily news. Exhibitors are responsible for bringing all hardware and software required for their demonstrations. Accepted exhibitors have to register to the conference like all the other conference participants. Please submit your proposal to: KDNet Office Ina Lauth ------------------------------------------------------ Codrina Lauth MA Fraunhofer Institute for Autonomous Intelligent Systems (AIS), KD - Knowledge Discovery Team - Schloss Birlinghoven D-53754 Sankt Augustin, Germany Tel.: +49 - (0) 22 41 - 14 - 2686 Fax: +49 - (0) 22 41 - 14 - 2072 Mobile:+49 - (0)174 - 64 96 833 mailto:codrina.lauth at ais.fhg.de URL: http://www.ais.fraunhofer.de/KD/lauth.htm Institute: http://www.ais.fraunhofer.de Project: http://www.kdnet.org From mcgl-ci0 at wpmail.paisley.ac.uk Sat May 15 07:45:21 2004 From: mcgl-ci0 at wpmail.paisley.ac.uk (Stephen McGlinchey) Date: Sat, 15 May 2004 12:45:21 +0100 Subject: CFP - Special Session on Neural Networks in Games and Simulations Message-ID: C A L L F O R P A P E R S Special Session On NEURAL NETWORKS IN COMPUTER GAMES AND SIMULATION as part of International Conference on Computer Games: Artificial Intelligence, Design and Education (CGAIDE) 8-10 November 2004 Hosted at the Microsoft Campus, Reading, UK http://www.scit.wlv.ac.uk/~cm1822/cgaide.htm The main objectives of this special session are: * To provide a forum for presenting and discussing the application of neural networks in computer games and simulations * To promote collaboration between the computer games and computational intelligence research communities. Neural Networks (NN) in in Computer Games and Simulations ============================We invite submission of papers presenting recent research on Artificial Neural Networks applied to any aspect of computer games technology or simulation. The scope of the session may include topics from the following non-exclusive list. Autonomous control of game agents. Processing of game content data. Real-time feature extraction from game data. Simulation or anaylisis of emotions / intelligence. Automatic analysis of human player behaviours. Synthesis of dynamic game content. Strategic game applications (e.g. Go and Chess) Optimising search spaces. Submissions are encouraged from any other related application of neural networks in games or simulation. Full papers should be sent in Adobe pdf or MS Word format to stephen.mcglinchey at paisley.ac.uk by the submission deadline. Send a full paper - not previously published - in Word format (no smaller than 10-point format) to include author, address, e-mail, keywords and abstract (200 words max) in one of the following categories: Regular paper max. 5 pages Extended Paper max. 8 pages Critical Review paper max. 10 pages Important dates: Draft paper submission : 30 July 2004 Notification of acceptance : 23 August 2004 Camera-ready submission deadline : 13 September 2004 Please send enquiries to: Stephen McGlinchey (stephen.mcglinchey at paisley.ac.uk) School of Computing University of Paisley Paisley, Scotland or Darryl Charles (dk.charles at ulster.ac.uk) School of Computing and Information Engineering University of Ulster Coleraine, Northern Ireland. From Angela.Jakary at med.va.gov Sat May 15 18:48:26 2004 From: Angela.Jakary at med.va.gov (Angela.Jakary@med.va.gov) Date: Sat, 15 May 2004 17:48:26 -0500 Subject: Volunteer wanted to analyze psychiatric neuroimaging lab data. Po tential for NIH R21 grant collaboration! Message-ID: <6884EB498415D411A71A0000F803A4680A2C3A8E@VHASFCEXC1> Research Opportunity: Neural Networks Analysis of Neuroimaging Data Our psychiatric neuroimaging laboratory located in the San Francisco VA Medical Center is looking for an individual to apply neural network pattern recognition methods to our neuroimaging data (we have NeuralWare software but are open to suggestions). The ultimate goal is to discover new relationships between neuroimaging abnormalities--based on magnetic resonance imaging (MRI) tissue volume and proton magnetic resonance spectroscopic imaging (MRSI) metabolite measurements--and demographic/clinical data for individuals suffering from bipolar disorder, depression, or schizophrenia. It is hypothesized that the resultant relationships could suggest new, more accurate classification systems for these disorders. Knowledge of quantitative MRI and MRSI techniques is ideal, however training is available. This is an outstanding research opportunity for the right individual, and could evolve into grant-funded research. Specifically, the NIMH is interested in seeing new methods of data analysis applied to neuroimaging data, and we would therefore like to develop a competitive R21 grant proposal (no pilot data required) involving neural networks analysis of our data. If you're interested, please send an e-mail to angela.jakary at med.va.gov. Thanks. From matsuka at psychology.rutgers.edu Wed May 19 20:53:41 2004 From: matsuka at psychology.rutgers.edu (Matsuka, Toshihiko) Date: Wed, 19 May 2004 20:53:41 -0400 Subject: [Paper available] COMBINATORIAL CODES in Obj Recognition Message-ID: <004e01c43e04$e5149860$0d5fe6a5@mp> Dear Connectionists: We would like to announce the paper: COMBINATORIAL CODES IN VENTRAL TEMPORAL LOBE FOR OBJECT RECOGNITION: HAXBY (2001) REVISITED: IS THERE A "FACE" AREA? by Stephen Jose Hanson, Toshihiko Matsuka, and James V. Haxby. to appear in NeuroImage ---abstract--- Haxby et al (2001) recently argued that category-related responses in the ventral temporal (VT) lobe during visual object identification were overlapping and distributed in topography. This observation contrasts with prevailing views that object codes are focal and localized to specific areas such as the fusiform and parahippocampal gyri. We provide a critical test of Haxby's hypothesis using a Neural Network classifier which can detect more general topographic representations and achieves 83% correct generalization performance on patterns of voxel responses in out-of-sample tests. Using voxel-wise sensitivity analysis we show that substantially the same VT lobe voxels contribute to the classification of all object categories, suggesting the code is combinatorial. Moreover, we found no evidence for local single category representations. The neural network representations of the voxel codes were sensitive to both category and to superordinate level features which were only available implicitly in the object categories. A preprint is available at: http://www.rumba.rutgers.edu/pubs/objrecog_shj_tm_jvh.PDF Best regards, Toshihiko Matsuka ------------------ toshihiko matsuka rutgers university ------------------ From b.graham at cs.stir.ac.uk Thu May 20 12:46:14 2004 From: b.graham at cs.stir.ac.uk (Bruce Philip Graham) Date: Thu, 20 May 2004 17:46:14 +0100 (BST) Subject: FINAL CALL: Edinburgh Summer School in Neuroinformatics Simulation Tools] Message-ID: <2592.139.153.254.219.1085071574.squirrel@yen.cs.stir.ac.uk> 2004 EDINBURGH SUMMER SCHOOL IN NEUROINFORMATICS SIMULATION TOOLS Institute for Adaptive and Neural Computation University of Edinburgh. August 23-27, 2004 FINAL CALL: * Application deadline: 4th June 2004 * This summer school in Neuroinformatics Simulation Tools is a five day intensive course which will provide a practical introduction to using neuroscience simulation tools including NEURON and Catacomb for computational modelling of neural systems. The course will be held at the e-Science Institute in the centre of Edinburgh and will include talks from invited speakers as well as practical hands-on experience using the latest simulation tools. The course is aimed at PhD students, postdocs and faculty in neuroscience and related disciplines wishing to learn how to apply computational modelling techniques to their research problems. Bursaries are available for PhD students to help with travel and accommodation costs. For more details, and an online application form visit:- http://www.anc.ed.ac.uk/school -- Dr Bruce Graham, Lecturer (b.graham at cs.stir.ac.uk) Dept. of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA phone: +44 1786 467 432 fax: +44 1786 464 551 -- The University of Stirling is a university established in Scotland by charter at Stirling, FK9 4LA. Privileged/Confidential Information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), you may not disclose, copy or deliver this message to anyone and any action taken or omitted to be taken in reliance on it, is prohibited and may be unlawful. In such case, you should destroy this message and kindly notify the sender by reply email. Please advise immediately if you or your employer do not consent to Internet email for messages of this kind. From ltbip at gatsby.ucl.ac.uk Fri May 21 09:32:40 2004 From: ltbip at gatsby.ucl.ac.uk (ltbip@gatsby.ucl.ac.uk) Date: Fri, 21 May 2004 14:32:40 +0100 Subject: Workshop on Learning Theoretic and Bayesian Inductive Principles Message-ID: <200405211332.i4LDWekX021180@wald.gatsby.ucl.ac.uk> Call for Participation EU PASCAL Workshop on Learning Theoretic and Bayesian Inductive Principles 19-21 July 2004 Gatsby Unit University College London, UK http://www.gatsby.ucl.ac.uk/~zoubin/LTBIP The theoretical analysis of systems that learn from data has been an important topic of study in statistics, machine learning, and information theory. Formal tools for this analysis have been developed from distinct inductive paradigms, such as Bayesian inference, statistical learning theory, and minimum description length. In recent years, there has been a convergence of ideas from these distinct paradigms, an example of which are PAC-Bayesian bounds on generalisation performance. The goal of this workshop is to bring together leading theoreticians to allow them to debate, compare and cross-fertilise ideas from these distinct inductive principles. The format of the workshop will be 90 minute talks from invited speakers, short talks from contributed speakers, and ample discussion time. We invite original contributions to the workshop that will undergo a short refereeing process. Please describe the work in 1-4 pages specifying the model that you adopt and outlining the results/contribution. We encourage blue-skies ideas but they should be sufficiently concrete to be clear and understandable. The deadline for contributions is 31 May 2004. IMPORTANT DATES: Deadline for Contributions: 31 May 2004 Notification of Acceptance: 14 Jun 2004 Registration Deadline(*): 28 Jun 2004 Workshop Dates: 19-21 Jul 2004 INVITED SPEAKERS: Peter Bartlett, University of California at Berkeley, USA Olivier Catoni, Universite Paris 6, France A Philip Dawid, University College London, UK Claudio Gentile, Universita dell'Insubria, Italy Gabor Lugosi, Pompeu Fabra University, Spain John Langford, Toyota Technological Institute, USA David MacKay, University of Cambridge, UK Paul Vitanyi, CWI, Netherlands ORGANISERS: Zoubin Ghahramani, University College London, UK John Shawe-Taylor, University of Southampton, UK PROGRAMME COMMITTEE: Zoubin Ghahramani, University College London, UK Peter Grnwald, CWI, Netherlands John Langford, Toyota Technological Institute, USA Gabor Lugosi, Pompeu Fabra University, Spain Shahar Mendelson, Australian National University, Australia John Shawe-Taylor, University of Southampton, UK ---------------------------------------------------------------------- PASCAL (Pattern Analysis, Statistical Modelling and Computational Learning, http:///www.pascal-network.org) is a newly launched (December 2003) European Network of Excellence (NoE) as part of its IST program. The NoE brings together experts from basic research areas such as Statistics, Optimisation and Computational Learning and from a number of application areas, with the objective of integrating research agendas and improving the state of the art in all concerned fields with emphasis on applications in multi-modal interfaces. From chiba at Cogsci.ucsd.edu Sat May 22 10:00:44 2004 From: chiba at Cogsci.ucsd.edu (Andrea Chiba) Date: Sat, 22 May 2004 07:00:44 -0700 (PDT) Subject: ICDL 04 Deadline Extended May 23 In-Reply-To: Message-ID: Due to unexpected problems with the submission site, papers and abstracts are still being accepted through Sunday, May 23rd. > ICDL 2004 FINAL Call for papers > DEADLINE IS SUNDAY, MAY 23 2004 > There will be absolutely no further extensions. > > THIRD INTERNATIONAL CONFERENCE ON DEVELOPMENT AND LEARNING: DEVELOPING > SOCIAL BRAINS > > The Salk Institute > October 20-22, 2004 > San Diego, California > A Satellite Conference preceding > The Annual Society for Neuroscience Conference > http://www.icdl.cc > > The goal of the conference is to bring together leading researchers in > neuroscience, machine learning, robotics, and developmental psychology, in > order to gain new insights about learning and development in natural > organisms and robots. The scope of developmental processes to be > considered is broad, including cognitive, social, emotional, and many > other skills exhibited by humans, and animals. The theme of the conference > this year will be "Developing Social Brains", but other topics related to > development and learning are welcome. > > PAPER SUBMISSION > The extended submission deadline is May 21, 2004. Papers for the meeting > can be submitted ONLY through the conference's web site at: > http://www.icdl.cc. Papers can be submitted either as a 200 word summary > or as a full paper (max 8 typeset pages). > IMPORTANT: There will be NO further submission deadline extensions. > > SPECIAL ISSUE ON NEUROCOMPUTING > Authors of selected papers will be invited to submit an extended version > of their paper for publication in a special issue of the Neurocomputing > Journal, on Development, Learning, and the Social Brain, published by > Elsevier Science B.V. > (http://www.elsevier.nl/locate/neucom) > > INVITED TALKS (Not yet confirmed) > John Allman > Dana Ballard > Rodney Brooks > Eric Courchesne > Peter Dayan > Jeff Elman > William Greenough > James L. McClelland > Pietro Perona > Terrence Sejnowski > Joan Stiles > John Watson > > REVIEW PROCESS > All submitted papers will be reviewed by the program committee. Papers > will be judged and accepted for the meeting based on the clarity with > which the work is described and the relevance to the goals of the > conference. All accepted papers not selected for oral talks as well as > papers explicitly submitted as poster presentations will be included in > one of three evening poster sessions. Authors will be notified of the > presentation format of their papers by the beginning of August. > > ORGANIZING COMMITTEE: > General Chair: Javier R. Movellan: > Co-Chairs: Andrea Chiba, Gedeon Deak, Jochen Triesch. > Program Chair: Jochen Triesch and Tony Jebara. > Program Co-Chairs: Marian Stewart-Bartlett, Gwen Ford Littlewort. > Publications Chair: Gedeon Deak. > > ADVISORY BOARD: > Jeff Elman > Pat Langley > James L. McClelland > Sandy Pentland > Terrence Sejnowski > Mriganka Sur > Esther Thelen > Juyang Weng > > PROGRAM COMMITTEE: > Minoru Asada > Dana Ballard > Luis Baumela > Simon Baron-Cohen > Mark Baxter > Leslie Carver > Jeff Cohn > Kerstin Dautenhahn > Kenji Doya > Martha Farah > Teresa Farroni > Masahiro Fujita > Ann Graybiel > William Greenough > Michael Hasselmo > Shoji Itakura > Hiroshi Ishiguro > Robert Jacobs > David Kleinfeld > Mark Konishi > Kang Lee > Denis Mareshal > Risto Miikulainen > Douglas Nitz > Roz Picard > Steven Quartz > Rajesh Rao > Matthew Schlesinger > Gregor Schoener > Geoffrey Schoenbaum > Linda Smith > Olaf Sporns > Luc Steels > Valerie Stone > Manuela Veloso > Paul Verschure > Christoph von der Malsburg > Hiroyuki Yano > > CONFERENCE REGISTRATION: > Register online at http://www.icdl.cc > Student Registration is: $150 > Non-student Registration is: $290 > > > From tere-ic0 at wpmail.paisley.ac.uk Fri May 21 12:27:56 2004 From: tere-ic0 at wpmail.paisley.ac.uk (Valery Tereshko) Date: Fri, 21 May 2004 17:27:56 +0100 Subject: PhD studentship Message-ID: Research studentship to work on Emergent Collective Behaviour in Ensembles of Communicating Agents is available in School of Computing of the University of Paisley. The approach is based on consideration of an agent ensemble as dynamical system and studying the latter with the methods of nonlinear dynamics and statistical physics. Knowledge of basic nonlinear dynamics and good computational skill are essential. Experience in and knowledge of statistical physics, chaos theory, neural and biologically-inspired computations, and swarm intelligence are desirable. Contact: Valery Tereshko, School of Computing, University of Paisley, Paisley PA1 2BE. e-mail: valery.tereshko at paisley.ac.uk From b.graham at cs.stir.ac.uk Thu May 20 12:42:56 2004 From: b.graham at cs.stir.ac.uk (Bruce Philip Graham) Date: Thu, 20 May 2004 17:42:56 +0100 (BST) Subject: FINAL CALL: Edinburgh Summer School in Neuroinformatics Simulation Tools] Message-ID: <2574.139.153.254.219.1085071376.squirrel@yen.cs.stir.ac.uk> 2004 EDINBURGH SUMMER SCHOOL IN NEUROINFORMATICS SIMULATION TOOLS Institute for Adaptive and Neural Computation University of Edinburgh. August 23-27, 2004 FINAL CALL:* Application deadline: 4th June 2004 * This summer school in Neuroinformatics Simulation Tools is a five day intensive course which will provide a practical introduction to using neuroscience simulation tools including NEURON and Catacomb for computational modelling of neural systems. The course will be held at the e-Science Institute in the centre of Edinburgh and will include talks from invited speakers as well as practical hands-on experience using the latest simulation tools. The course is aimed at PhD students, postdocs and faculty in neuroscience and related disciplines wishing to learn how to apply computational modelling techniques to their research problems. Bursaries are available for PhD students to help with travel and accommodation costs. For more details, and an online application form visit:- http://www.anc.ed.ac.uk/school -- Dr Bruce Graham, Lecturer (b.graham at cs.stir.ac.uk) Dept. of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA phone: +44 1786 467 432 fax: +44 1786 464 551 From terry at salk.edu Mon May 24 18:48:33 2004 From: terry at salk.edu (Terry Sejnowski) Date: Mon, 24 May 2004 15:48:33 -0700 (PDT) Subject: NEURAL COMPUTATION 16:7 In-Reply-To: <200405121829.i4CITNt96552@dax.snl.salk.edu> Message-ID: <200405242248.i4OMmXL28618@dax.snl.salk.edu> Neural Computation - Contents - Volume 16, Number 7 - July 1, 2004 NOTES Bayesian Estimation of Stimulus Responses in Poisson Spike Trains Sidney R. Lehky Comments on "A Parallel Mixture of SVMs for Very Large Scale Problems" Xiaomei Liu, Lawrence O. Hall, and Kevin W. Bowyer LETTERS Automated Algorithms for Multiscale Morphometry of Neuronal Dendrites Christina M. Weaver, Patrick R. Hof, Susan L. Wearne, and W. Brent Lindquist Computing and Stability in Cortical Networks Peter E. Latham and Sheila Nirenberg Real-Time Computation at the Edge of Chaos in Recurrent Neural Networks Nils Bertschinger and Thomas Natschlaeger Information Geometry of U-Boost and Bregman Divergence Noboru Murata, Takashi Takenouchi, Takafumi Kanamori and Shinto Eguchi A New Approach to the Extraction of ANN Rules and Their Generalization Capacity Through GP Juan R. Rabunal, Julian Dorado, Alejandro Pazos, Javier Pereira and Daniel Rivero A Classification Paradigm for Distributed Vertically Partitioned Data Jayanta Basak and Ravi Kothari ----- 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 michael.spratling at kcl.ac.uk Tue May 25 04:54:29 2004 From: michael.spratling at kcl.ac.uk (Michael Spratling) Date: Tue, 25 May 2004 09:54:29 +0100 Subject: Post-doc position, King's College London Message-ID: <40B309C5.3060708@kcl.ac.uk> Postdoctoral Research Associate King's College London 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 23,144 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 auke.ijspeert at epfl.ch Tue May 25 08:09:13 2004 From: auke.ijspeert at epfl.ch (Auke Jan Ijspeert) Date: Tue, 25 May 2004 14:09:13 +0200 Subject: SAB2004: call for abstracts and early registration deadlines (May 28 and 31) Message-ID: <40B33769.3080709@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. Best regards, Auke Ijspeert ----- SAB2004 Call for participation, Call for Abstracts ----- 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 ----- We are glad to announce that the registration for SAB2004 is now open, and that the preliminary technical program is available. The deadline for early registration is 28 May 2004. See http://www.isab.org/sab04/ for details. ----- Call for abstracts ----- SAB2004 introduces a new feature: the Last-Minute-Results poster session. 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. Deadlines: May 31 2004: Abstract submission deadline June 7 2004: Notification of acceptance 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 From d.g.heinke at bham.ac.uk Tue May 25 11:33:09 2004 From: d.g.heinke at bham.ac.uk (Dietmar Heinke) Date: Tue, 25 May 2004 16:33:09 +0100 Subject: Position in Computational Neuroscience Message-ID: <40B36735.5000706@bham.ac.uk> Research Fellow/Assistant in Computational Neuroscience. Behavioural Brain Sciences Centre School of Psychology University of Birmingham Birmingham The Fellow/Assistant will work with Dr Dietmar Heinke and Prof Glyn Humphreys in developing a computational model of visual object recognition and attention in humans. The framework for this project is set by the computational model, SAIM (Selective Attention for Identification Model; see http://web.bham.ac.uk/heinkedg/SAIM for details) and will draw on experimental evidence from a broad range of sources: single cell studies, fMRI, MR tractography, neuropsychological and psychological studies. The position is funded by the BBSRC and is part of the collaboration of the following four groups: Glyn Humphreys and Dietmar Heinke, Behavioural Brain Sciences Centre, Birmingham; S. Shipp, Institute of Cognitive Neuroscience, University College London; Paul Matthews and colleagues, Oxford Centre for Functional Magnetic Resonance Imaging of the Brain and Andrew T. Smith, Royal Holloway & Bedford New College. The project is part of a long-term effort in the Behavioural Brain Sciences Centre (BBS) to establish computational modelling as a standard method within cognitive neuroscience. Examples for first results can be found on Dietmar Heinke's web-page and the BBS web-page. This long-term project is also supported by long-standing collaborations with the school of computer science, including Aaron Sloman, Ela Claridge and John Bullinaria Applicants should have, or be about to complete, a PhD or a very good MSc in a relevant field (incl. computer vision, neural networks, connectionist modelling, physics, or applied mathematics). Familiarity with programming (MatLab, C++) would be an advantage. The posts are available from 1 September 2004 or as soon as possible thereafter. Application forms are available from Personnel Services, University of Birmingham (j.s.thorp at bham.ac.uk) and informal queries should be addressed to Dietmar Heinke (0121 414 4920 d.g.heinke at bham.ac.uk). The salary is at point 6 on the RA1A scale (18,893 to 21,010). The deadline is 30 June 2004. -- School of Psychology University of Birmingham Birmingham B15 2TT, UK http://web.bham.ac.uk/heinkedg/ Phone: +44 121-414-4920 FAX: +44 121-414-4897 From geoff at cns.georgetown.edu Wed May 26 09:46:50 2004 From: geoff at cns.georgetown.edu (Geoff Goodhill) Date: Wed, 26 May 2004 09:46:50 -0400 Subject: Papers on axon guidance available Message-ID: Dear Connectionists, The following two recent papers may be of interest to readers of this list. They can be downloaded from http://cns.georgetown.edu/~geoff/pubs.html Rosoff, W.J., Urbach, J.S., Esrick, M., McAllister, R.G., Richards, L.J. & Goodhill, G.J. (2004). A new chemotaxis assay shows the extreme sensitivity of axons to molecular gradients. Nature Neuroscience, 7, 678-682. Goodhill, G.J., Gu, M. & Urbach, J.S. (2004). Predicting axonal response to molecular gradients with a computational model of filopodial dynamics. Neural Computation, in press. Abstracts are appended below. Sincerely, Geoff Geoffrey J Goodhill, PhD Department of Neuroscience Georgetown University Medical Center 3900 Reservoir Road NW, Washington DC 20007 Email: geoff at georgetown.edu Homepage: cns.georgetown.edu ------------ A new chemotaxis assay shows the extreme sensitivity of axons to molecular gradients. Axonal chemotaxis is believed to play a key role in wiring up the developing and regenerating nervous system, but little is known about how axons actually respond to molecular gradients. We report a new quantitative assay that allows the long-term response of axons to gradients of known and controllable shape to be examined in a three-dimensional gel. Using this assay we show that axons may be nature's most sensitive gradient detectors, but that this sensitivity exists only within a narrow range of ligand concentrations. This assay should also be applicable to other biological processes controlled by molecular gradients, such as cell migration and morphogenesis. ------------ Predicting axonal response to molecular gradients with a computational model of filopodial dynamics. Axons are often guided to their targets in the developing nervous system by attractive or repulsive molecular concentration gradients. We propose a computational model for gradient sensing and directed movement of the growth cone mediated by filopodia. We show that relatively simple mechanisms are sufficient to generate realistic trajectories for both the short term response of axons to steep gradients and the long term response of axons to shallow gradients. The model makes testable predictions for axonal response to attractive and repulsive gradients of different concentrations and steepness, the size of the intracellular amplification of the gradient signal, and the differences in intracellular signaling required for repulsive versus attractive turning. From J.L.Wyatt at cs.bham.ac.uk Wed May 26 09:56:21 2004 From: J.L.Wyatt at cs.bham.ac.uk (Jeremy L Wyatt) Date: Wed, 26 May 2004 14:56:21 +0100 (BST) Subject: looking for two outstanding post-docs for an ambitious EC cognitive systems project Message-ID: Dear colleague, I'm writing to a long list of people because we are looking for two outstanding postdocs to work on a large EU funded project called CoSy. The project involves seven sites around the EU, including people working in natural language, robotics, vision, learning, psychology and AI. The aim is to put together many things that are normally done separately, e.g. vision, language, planning, plan execution, 3-D manipulation, learning, self-understanding, etc. integrated in a working robot in a coherent architecture. The project is described in greater detail here: http://www.cs.bham.ac.uk/research/projects/cosy/ Because of the breadth of the project we hope to find people with broad and deep knowledge of AI, at least one of whom should have experience in robotics. Because the tasks are so challenging we have managed to obtain funding for higher salaries than usual (at least in the UK) for post-doctoral positions. The Birmingham portion of the project is led by Aaron Sloman and myself. If you know of any potential candidates please pass on the appended advertisement directly, or circulate within your group. If you are interested then some of the work in the Birmingham part of the project is described here: http://www.cs.bham.ac.uk/research/projects/cosy/PlayMate-start.html If you have time, please feel free to comment, criticise, join in, etc. We hope that the project will be as open as possible, including the sharing of code, as well as ideas. The advert follows. cheers Jeremy PS Aaron Sloman is also circulating the advert in a similar way. While we've tried to make our mailing lists disjoint, I'd like to apologise if you've seen this advert more than once. -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Dr Jeremy L Wyatt Intelligent Robotics Laboratory Tel 00 44 121 414 4788 School of Computer Science Fax 00 44 121 414 4281 University of Birmingham jlw at cs.bham.ac.uk Birmingham, B15 2TT, UK www.cs.bham.ac.uk/~jlw www.cs.bham.ac.uk/research/robotics -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= ________________ Subject: Two Research Posts Birmingham UK: EC Cognitive Systems Robotic project THE UNIVERSITY OF BIRMINGHAM SCHOOL OF COMPUTER SCIENCE http://www.cs.bham.ac.uk/ Research Fellows (2 Posts) on EC-Funded 'Cognitive Systems' Integrated Project Applications are invited for two experienced researchers to work on an ambitious seven partner EC-funded project (CoSy: Cognitive Systems for Cognitive Assistants). The whole project will involve upward of twenty people including partners working in robotics, natural language, experimental and developmental psychology, computer vision, machine learning and artificial intelligence. The work at Birmingham will involve designing and implementing mechanisms and a cognitive architecture for a robot that sees, manipulates and talks about objects on a table. Much of the emphasis at Birmingham will be on developing and implementing architectures that allow us to integrate the various components of an intelligent system. The aim of the whole project is to investigate ways of developing physically embodied robots that combine perception, learning, reasoning, planning, acting, communication and self-awareness in an integrated architecture. One of the posts will require some robotics experience. The remit of the project is, however, very broad so people with experience in any area of AI are encouraged to apply. Relevant areas include, but are not limited to: reasoning, representation, architectures, learning, natural language understanding, robotics, and computer vision. Further details of the project and a draft description of work to be done in Birmingham can be found at: http://www.cs.bham.ac.uk/research/projects/cosy/ A job specification with instructions for applicants is available at http://www.cs.bham.ac.uk/news/jobs/cosy.rf.04/ Informal enquiries can be made to the two Principal Investigators at Birmingham: Dr Jeremy Wyatt or Professor Aaron Sloman Starting salary: on scale 21,852 - 31,715 a year depending on experience and qualifications, with potential progression to 34,838 (Scales are currently under review). Likely starting date for the project is: 1 September 2004. Appointments will be for a period of 18 months initially, with an expected extension to 48 months (standard EC procedure). Application forms are returnable by 10th June 2004, though late applicants may be considered. Further enquiries and requests for application forms may be made to Personnel Services, The University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK. Tel: +44 121 415 9000 Email: Ms. Jo Gerald Web: www.punit.bham.ac.uk/vacancies Please quote reference S36732 in all communications. Working towards equal opportunities. ==== From oby at cs.tu-berlin.de Fri May 28 03:40:49 2004 From: oby at cs.tu-berlin.de (Klaus Obermayer) Date: Fri, 28 May 2004 09:40:49 +0200 (MEST) Subject: tenured faculty position Message-ID: Dear All, below please find an announcement for an open faculty position (tenured) in the area of artificial intelligence in our department (EE and CS) at the Technical University of Berlin. The Berlin area has a high concentration of high quality research institutions (three universities, Max-Planck and Fraunhofer institutes, etc.) and a lively machine learning and computational neuroscience scene. Besides that, it is a pleasant city to live in and Germany's most prominent place for cultural activities. Cheers Klaus ========================================================================== FACULTY POSITION IN ARTIFICIAL INTELLIGENCE Faculty IV, Technical University of Berlin, Berlin, Germany The Department for Electrical Engineering and Computer Science solicits application for a tenured faculty position (salary level C3) in the area of artificial intelligence. Potential areas of research include (but are not restricted to) machine learning, automatic deduction and problem solving, planning and decision making, knowledge representation, scene analysis, and speech processing. The successful candidate is expected to join the department's undergraduate teaching programs, as well as the graduate education in the area of Artificial Intelligence. The successful candidate is expected to teach courses in German after a few years. Requirements: Ph.D. degree and Habilitation or equivalent achievements (cf. =A7100 BerlHG); research experience in the field of artificial intelligence; a strong publication record and teaching experience. Experience in the acquisition of research grants is desirable. Please send applications to: Chair of the Department of Electrical Engineering and Computer Science Technical University of Berlin FR 5-1, Franklinstrasse 28/29 10587 Berlin, Germany email: fbv13 at cs.tu-berlin.de All applications received before June 25th, 2004, will be given full consideration. The Technical University of Berlin wants to increase the percentage of women on its faculty and strongly encourages applications from qualified individuals. Women will be preferred given equal qualifications. Handicapped persons will be preferred given equal qualifications. ========================================================================= Prof. Dr. Klaus Obermayer phone: 49-30-314-73442 FR2-1, NI, Informatik 49-30-314-73120 Technische Universitaet Berlin fax: 49-30-314-73121 Franklinstrasse 28/29 e-mail: oby at cs.tu-berlin.de 10587 Berlin, Germany http://ni.cs.tu-berlin.de/ From zhong at cse.fau.edu Fri May 28 15:32:19 2004 From: zhong at cse.fau.edu (zhong@cse.fau.edu) Date: Fri, 28 May 2004 15:32:19 -0400 (EDT) Subject: CFP: ICTAI-2004 special track on Unsupervised and Semi-supervised Learning Message-ID: <36833.192.35.232.241.1085772739.squirrel@webmail.cse.fau.edu> (sincere apologies if you received multiple copies of this notice) Dear Colleagues, We would like to bring to your attention a special track of ICTAI-2004 on unsupervised and semi-supervised learning. The 16th International Conference on Tools with Artificial Intelligence (ICTAI-2004) will be held in Boca Raton, Florida on November 15-17, 2004. The special track will be part of the conference with a focus on reporting recent advances in theories and applciations of unsupervised and semi-supervised learning techniques. The web site for the main conference is http://www.cse.fau.edu/~ictai04/ and the URL for the CFP of the special track is http://www.cse.fau.edu/~zhong/cfp-learning.htm Deadlines: June 18 - paper submission due August 2 - acceptance notification September 3 - final camera ready paper due Any questions please email the co-chairs of this special track at davidson at cs.albany.edu or zhong at cse.fau.edu. Also please feel free to distribute this notice to people who you think are interested. sincerely, Ian Davidson Shi Zhong Co-chairs ICTAI-2004 Special Track on Theories and Applications of Unsupervised and Semi-Supervised Learning From ash at cs.umass.edu Fri May 28 10:32:00 2004 From: ash at cs.umass.edu (Ashvin Shah) Date: Fri, 28 May 2004 10:32:00 -0400 (EDT) Subject: links to J Neurophys article: Cortical Involvement in the Recruitment of Wrist Muscles In-Reply-To: Message-ID: Hello Connectionists - the following article may be of interest. I've included the citation and the abstract. Thank you. - Ashvin Shah, ash at cs.umass.edu, (413) 545-1596 Neuroscience and Behavior Program, University of Massachusetts, Amherst Shah, A., Fagg, A. H., and Barto, A. G. (2004) Cortical Involvement in the Recruitment of Wrist Muscles Journal of Neurophysiology vol. 99(6), pages 2445 - 2456 http://www-all.cs.umass.edu/pubs/2004/shah_fb_JNP04.pdf http://www-all.cs.umass.edu/pubs/2004/shah_fb_JNP04.ps http://jn.physiology.org/ PMID: 14749314 [PubMed - in process] ABSTRACT: In executing a voluntary movement, one is faced with the problem of translating a specification of the movement in task space (e.g., a visual goal) into a muscle-recruitment pattern. Among many brain regions, the primary motor cortex (MI) plays a prominent role in the specification of movements. In what coordinate frame MI represents movement has been a topic of considerable debate. In a two-dimensional wrist step-tracking experiment, Kakei et al. described some MI cells as encoding movement in a muscle-coordinate frame and other cells as encoding movement in an extrinsic-coordinate frame. This result was interpreted as evidence for a cascade of transformations within MI from an extrinsic representation of movement to a muscle-like representation. However, we present a model that demonstrates that, given a realistic extrinsic-like representation of movement, a simple linear network is capable of representing the transformation from an extrinsic space to the muscle-recruitment patterns implementing the movements on which Kakei et al. focused. This suggests that cells exhibiting extrinsic-like qualities can be involved in the direct recruitment of spinal motor neurons. These results call into question models that presume a serial cascade of transformations terminating with MI pyramidal tract neurons that vary their activation exclusively with muscle activity. Further analysis of the model shows that the correlation between the activity of an MI neuron and a muscle does not predict the strength of the connection between the MI neuron and muscle. This result cautions against the use of correlation methods as a measure of cellular connectivity. -- From poznan at iub-psych.psych.indiana.edu Fri May 28 15:43:36 2004 From: poznan at iub-psych.psych.indiana.edu (Roman R. Poznanski) Date: Fri, 28 May 2004 14:43:36 -0500 Subject: A new book from CRC Press on Neuromimetic robotics Message-ID: <40B79668.8080804@iub-psych.psych.indiana.edu> MODELING IN THE NEUROSCIENCES: FROM BIOLOGICAL SYSTEMS TO NEUROMIMETIC ROBOTICS SECOND EDITION Edited by G.N.Reeke, R.R. Poznanski, K.A. Lindsay, J.R. Rosenberg, and O.Sporns The Second Edition is a thoroughly updated and carefully written and edited volume, dedicated to covering recent conceptual advances in the neurosciences as well as the basics of the field. It is essential for those interested in theoretical neuroscience and its implementation via integrative and synthetic neuronal modeling. Many state-of-the-art examples are presented to show how mathematical and computer modeling can contribute at various levels in the hierarchy of the neurosciences. Each chapter includes a problem set. Data sets are available online to add coverage for future advances in this rapidly changing and expanding field. This seminal book develops models that provide insights into underlying physiological mechanisms at the molecular, cellular, and network levels. Physiological mechanisms, such as voltage-dependent ion channels, calcium-dependent release, neurotransmitter-receptor dynamics and second-messenger effects, backpropagating sodium spikes, as well as cable properties governing electric current flow in neurons and neuronal networks are discussed throughout the book. The final chapters outline directions of theoretical modeling leading towards neural robotics and the fabrication of neuromimetic robotic devices beyond what is currently available in today's robotic technology. Dimensions Hardcover (grey vinyl board with gold lettering on light green spine) + jacket cover (maroon red cover with golden green micrograph superimposed on black background), Approx. 740 pages, 280 illus., Bibliography :-1600 refs. Further information: www.crcpress.com . Sale Price: US$139.95 --------------070505030109010106050503 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit
MODELING IN THE NEUROSCIENCES: FROM BIOLOGICAL SYSTEMS TO NEUROMIMETIC ROBOTICS
SECOND EDITION

Edited by
G.N.Reeke, R.R. Poznanski, K.A. Lindsay, J.R. Rosenberg, and O.Sporns

The Second Edition is a thoroughly updated and carefully written and edited
volume, dedicated to covering recent conceptual advances in the neurosciences
as well as the basics of the field. It is essential for those interested in
theoretical neuroscience and its implementation via integrative and synthetic
neuronal modeling. Many state-of-the-art examples are presented to show how
mathematical and computer modeling can contribute at various levels in the
hierarchy of the neurosciences. Each chapter includes a problem set. Data
sets are available online to add coverage for future advances in this rapidly
changing and expanding field.
This seminal book develops models that provide insights into underlying
physiological mechanisms at the molecular, cellular, and network levels.
Physiological mechanisms, such as voltage-dependent ion channels,
calcium-dependent release, neurotransmitter-receptor dynamics and 
second-messenger effects, backpropagating sodium spikes, as well as cable
properties governing electric current flow in neurons and neuronal networks
are discussed throughout the book. The final chapters outline directions of
theoretical modeling leading towards neural robotics and the fabrication of
neuromimetic robotic devices beyond what is currently available in
today's robotic technology.

Dimensions

Hardcover (grey vinyl board with gold lettering on light green spine) +
jacket cover (maroon red cover with golden green micrograph superimposed on
black background), Approx. 740 pages, 280 illus., Bibliography :-1600 refs.

Further information:  www.crcpress.com.
Sale Price: US$139.95







--------------070505030109010106050503-- From d.mandic at imperial.ac.uk Fri May 28 07:04:54 2004 From: d.mandic at imperial.ac.uk (Danilo P. Mandic) Date: Fri, 28 May 2004 12:04:54 +0100 Subject: Signal characterisation - TR and article Message-ID: <037901c444a3$9b2f1a90$5a7dc69b@MandicLaptop> Dear Connectionists may I draw your attention to the Technical Report entitled: On the Characterisation of the Deterministic/Stochastic and Linear/Nonlinear Nature of Time Series (T. Gautama, M. M. Van Hulle and D. P. Mandic)=20 It can be downloaded from http://www.commsp.ee.ic.ac.uk/~mandic It combines the concept of surrogate data and signal representation in phase space to provide an insight into the nature of a signal. We have applied it to fMRI, HRV, and EEG signals and it works fine. A more rigorous analysis can be found in the following Physica D article: Temujin Gautama, Danilo P. Mandic and Marc M. Van Hulle ' The delay vector variance method for detecting determinism and nonlinearity in time series ' Physica D: Nonlinear Phenomena, 190 (2004) 167-176=20 which can be downloaded from: http://134.58.34.50/temu/physD04_preprint.pdf or http://www.commsp.ee.ic.ac.uk/~mandic=20 The reasoning behind this analysis is that the recent progress in biomedicine has brought to light problems where an insight into the nature of a signal is necessary before performing the actual signal processing. This particularly applies to the linear vs nonlinear and deterministic vs stochastic signal properties (remember the Wold decomposition theorem?) The MatLab code which accompanies the articles can be downloaded from: DVV-toolbox: http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=3264&objectType=file and Surrogates toolbox: http://www.mathworks.com/matlabcentral/fileexchange/loadFile.do?objectId=4612&objectType=FILE Your comments are most welcome. Enjoy! Danilo ============================= Dr Danilo Mandic Senior Lecturer in Signal Processing Department of Electrical and Electronic Engineering Imperial College London United Kingdom Phone: +44 207 594 6271 Fax: +44 207 594 6234 E-mail: d.mandic at imperial.ac.uk http://www.commsp.ee.ic.ac.uk/~mandic From nips04pub at hotmail.com Mon May 31 16:42:33 2004 From: nips04pub at hotmail.com (John Platt) Date: Mon, 31 May 2004 13:42:33 -0700 Subject: NIPS 2004 Call for Workshops Message-ID: CALL FOR WORKSHOPS --- NIPS 2004 Neural Information Processing Systems --- Natural and Synthetic Friday, December 17 --- Saturday, December 18, 2004 Vancouver, British Columbia, Canada http://nips.cc Deadline for Workshop Proposals: August 1, 2004 Following the regular program of the Neural Information Processing Systems 2004 Conference in Vancouver, BC, Canada, workshops on various current topics in Neural Information Processing will be held on December 17 and 18, 2004, in Whistler, BC, Canada. We invite researchers interested in chairing one of these workshops to submit proposals for possible workshops. The goal of the workshops is to provide an informal forum for researchers to discuss important research questions and challenges. Controversial issues, open problems, and comparisons of competing approaches are encouraged and preferred as workshop topics. Representation of alternative viewpoints and panel-style discussions are also particularly encouraged. Workshop topics include, but are not limited to, the following: Active Learning, Attention, Audition, Bayesian Networks, Bayesian Statistics, Benchmarking, Bioinformatics, Brain Imaging, Computational Complexity, Control, Genetic Algorithms, Graphical Models, Hippocampus and Memory, Human-Computer Interfaces, Hybrid Supervised/Unsupervised Learning, Implementations, Independent Component Analysis, Kernel Methods, Mean-Field Methods, Music, Network Dynamics, Neural Coding, Neural Plasticity, On-Line Learning, Optimization, Recurrent Nets, Robotics, Rule Extraction, Self-Organization, Sensory Biophysics, Signal Processing, Spike Timing, Speech, Time Series, Topological Maps, and Vision. Detailed descriptions of previous workshops may be found at: http://nips.cc under Previous Conferences. There will be six hours of workshop meetings per day, split into morning and afternoon sessions, with free time between the sessions for ongoing individual exchange or outdoor activities. Selected workshops may be invited to submit proceedings for publication in the post-NIPS workshops monographs series published by the MIT Press. Workshop organizers have several responsibilities including: * Coordinating workshop participation and content, which includes arranging short informal presentations by experts, arranging for expert commentators to sit on a discussion panel, formulating a set of discussion topics, etc. * Moderating the discussion, and reporting its findings and conclusions to the group during the evening plenary sessions. * Writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. SUBMISSION INSTRUCTIONS: Interested parties must submit a proposal for a workshop via email by August 1,2004. Proposals should include a title, description of what the workshop is to address and accomplish, proposed workshop length (1 or 2 days), planned format (e.g., lectures, group discussions, panel discussion, combinations of the above, etc.), and proposed speakers. Names of potential invitees should be given where possible. Preference will be given to workshops that reserve a significant portion of time for open discussion or panel discussion, as opposed to a pure "mini-conference" format. An example format is: * Tutorial lecture providing background and introducing terminology relevant to the topic. * Discussion or panel presentation. * Short talks or panels alternating with discussion and question/answer sessions. * General discussion and wrap-up. We suggest that organizers allocate at least 50 percent of the workshop schedule to questions, discussion, and breaks. Past experience suggests that workshops otherwise degrade into mini-conferences as talks begin to run over. For the same reason, we strongly recommend that each workshop include no more than 12 talks per day. The proposal should state why the topic is of interest, why it should be discussed, and who the targeted group of participants is. It should also include a brief resume of the prospective workshop chair with a list of publications to establish scholarship in the field. We encourage workshops that build, continue, or arise from one or more workshops from previous years. Please mention any such connections. NIPS does not provide travel funding for workshop speakers. In the past, some workshops have sought and received funding from external sources to bring in outside speakers. In addition, the organizers of each accepted workshop can name up to four people (six people for 2-day workshops) to receive discounted registration for the workshop program. Submissions should include the name, address, email address, phone and fax numbers for all organizers. If there is more than one organizer, please designate one organizer as the primary contact. Proposals should be emailed as plain text to nips-prp at umich.edu (please do not use attachments, Word, postscript, html, or PDF files) NIPS 2004 WORKSHOP CO-CHAIRS: * Satinder Singh, University of Michigan * Daniel D. Lee, University of Pennsylvania Questions may be emailed to nips-adm at umich.edu DEADLINE FOR WORKSHOP PROPOSALS: AUGUST 1, 2004 REMINDER: Deadline for Papers --- June 4, 2004 From torsello at dsi.unive.it Mon May 31 08:59:37 2004 From: torsello at dsi.unive.it (Andrea Torsello) Date: Mon, 31 May 2004 14:59:37 +0200 Subject: Special Issue on Similarity-Based Pattern Recognition Message-ID: <200405311459.37577.torsello@dsi.unive.it> CALL FOR PAPERS PATTERN RECOGNITION Special Issue on Similarity-Based Pattern Recognition Submission deadline: December 15, 2004 Guest Editors: Manuele Bicego, University of Sassari, Italy (bicego at uniss.it) Vittorio Murino, University of Verona, Italy (vittorio.murino at univr.it) Marcello Pelillo, University of Venice, Italy (pelillo at dsi.unive.it) Andrea Torsello, University of Venice, Italy (atorsell at dsi.unive.it) Traditional Pattern Recognition techniques are centered around the notion of "feature". According to this view, the objects to be clustered or classified are represented in terms of properties that are intrinsic to the object itself. Hence, a typical pattern recognition system makes its "decisions" by simply looking at one or more feature vectors fed as input. The strength of this approach is that it can leverage a wide range of mathematical tools ranging from statistics, to geometry, to optimization techniques. However, in many real-world applications the objects are not naturally representable in terms of a vector of features. For example, graph representations lack a canonical order or correspondence between nodes. Furthermore, even if a vector mapping can be established, the vectors will be of variable length, hence would not belong to a single vector-space. On the other hand, quite often it is possible to obtain a measure of the similarity/dissimilarity of the objects to be classified. It is therefore tempting to design a pattern recognizer which, unlike traditional systems, accepts as input a matrix containing the similarities between objects and produces class labels as output. Indeed, recently, there has been a renewed interest in similarity-based techniques, both in developing and studying new effective distances between non vectorial entities, like graphs, sequences, structures, and in proposing alternative distance-based paradigms. These methods typically keep the algorithm generic and independent from the actual data representation, allowing the use of non-metric similarities (thereby violating the triangular inequality). Further, they make the approaches applicable to problems that do not have a natural embedding to a uniform feature space, such as the clustering of structural or graph-based representations or the analysis of sequences. Finally, these representations are well suited to both supervised and unsupervised classification. The literature of pairwise algorithms includes, among others, kernel methods, spectral clustering techniques/graph partitioning, quadratic optimization, self-organizing maps, etc. These techniques are successfully applied to very diverse problems like object classification, image retrieval by content, color quantization, image segmentation, perceptual grouping, and bioinformatics (gene alignment, gene classification or phylogenetic analysis). The goal of this special issue is to solicit and publish high-quality papers that bring a clear picture of the state of the art in this area. We aim to appeal to researchers in Pattern Recognition and Computer Vision who are using or developing similarity-based techniques. Papers are solicited that address theoretical as well as practical issues related to the Special Issue's theme. Topics of interest include (but are not limited to): - Similarity-based classification - Similarity-based clustering - Embeddings - Kernel methods - Spectral techniques - Definition and analysis of distances between sequences, structures, and images - Applications SUBMISSION PROCEDURE Only electronic submissions will be accepted. Manuscripts must be submitted via the special issue's web page http://www.dsi.unive.it/Similarity-PR/ The manuscripts must be submitted by December 15, 2004, and should conform to the standard guidelines of the Pattern Recognition journal. All submitted papers will be reviewed by at least three independent reviewers. -- Andrea Torsello PhD Dipartimento di Informatica, Universita' Ca' Foscari di Venezia via Torino 155, 30172 Venezia Mestre, Italy Tel: +39 0412348468 Fax: +39 0412348419 http://www.dsi.unive.it/~atorsell From lendasse at JAMES.HUT.FI Mon May 31 09:58:03 2004 From: lendasse at JAMES.HUT.FI (Amaury Lendasse) Date: Mon, 31 May 2004 16:58:03 +0300 Subject: Master Degree Thesis Position Message-ID: Dear Connectionists, I am looking for a Master Degree student. The position is described in http://www.cis.hut.fi/~lendasse/masterdegreeposition.html Dr. Amaury Lendasse *** PLEASE FORWARD TO POTENTIAL CANDIDATES *** MASTER DEGREE THESIS in the field of Time Series Prediction We propose a Master Degree Thesis in the field of Time series prediction and principally for the long-term prediction. These fields are briefly described below. This Master Degree Thesis should be done between September 2004 and December 2005. PROFILE: Applicants should have a Bachelor degree (or equivalent), in fields, such as Computer Science, Electrical Engineering, or Applied Math, or an equivalent or similar background. Knowledge about Neural Networks, Time Series Prediction, or System Identification is an advantage, but not a requirement. Ability to work in team is expected. The candidate will also have the following desirable qualifications: - strong programming skills (especially with Matlab); - good communication skills in English. APPLICATION FORM: Please send your application including a CV via E-mail to the contact person. CONTACT: Dr. Amaury Lendasse Helsinki University of Technology Laboratory of Computer and Information Science P.O. Box 5400 FIN-02015 HUT FINLAND tel. +358-9-451 4499 fax +358-9-451 3277 cell +358-40-770 0237 Email: lendasse at cis.hut.fi URL: http://www.cis.hut.fi/~lendasse Time series Prediction Time series forecasting is a challenge in many fields. In finance, one forecasts stock exchange indices or stock market indices; data processing specialists forecast the flow of information on their networks; producers of electricity forecast the load of the following day. The common point to their problems is the following: how can one analyze and use the past to predict the future? Many techniques exist: linear methods such as ARX, ARMA, etc., and nonlinear ones such as artificial neural networks. In general, these methods try to build a model of the process that is to be predicted. The model is then used on the last values of the series to predict future ones. The common difficulty to all methods is the determination of sufficient and necessary information for a good prediction. If the information is insufficient, the forecasting will be poor. On the contrary, if information is useless or redundant, modeling will be difficult or even skewed. Long-Term Prediction of Time Series As this problem can be found in many fields, many methods have been developed with very different approaches, from statistics to system identification and more recently neural networks. Most of the time, the models are linear and perform well on a rather short-term horizon, depending on the complexity of the problem. Their efficiency on a longer term is more questionable. This fact is due to the learning strategy used to fit the model to the data. The goal is usually to optimize the performance at a given term, most often the next time step. There are only a few attempts to explicitly predict values at long term, or at least global trends. This problem is quite hard since the uncertainty increases with the horizon of prediction. Another issue generally shared by classical models (such as ARX, ARMAX, .) is that they are used to predict a single value of a scalar time series. In practice some industrial applications require the prediction of a set of values in one single step instead of several independent values. Forecasting a vector of values requires more complex models able to predict several components together. If the approach is to develop several simple models and combine them to predict a vector, one can lose the correlation information between the vector components. Though each model may perform well, the forecasting accuracy could be rather poor when considering the vector of predicted values as a whole. Developing methods able to predict several values at each step, with the same expected performance on each value, should thus be a major concern. Let us consider now the general problem of forecasting at long term. Despite the fact that long-term predictions in real situations will probably never be very accurate, in some applications there is a need to have at least some ideas about the future of the time series. For example, answers to questions such as "Are there bounds on the future values?" or "What can we expect in average?" or even "Are the confidence intervals on future values large or narrow?" can give some ideas about the time series evolution at long-term. Publications about Times Predictions can be found in http://www.cis.hut.fi/~lendasse/Publications.html