From giro-ci0 at wpmail.paisley.ac.uk Mon Apr 1 17:09:42 2002 From: giro-ci0 at wpmail.paisley.ac.uk (Mark Girolami) Date: Mon, 01 Apr 2002 23:09:42 +0100 Subject: Research Positions Message-ID: Post-Doctoral Research Assistantship A research project which is funded by the Engineering and Physical Sciences Research Council (EPSRC), the Department of Trade and Industry (DTI), and industrial partners will be conducted at the University of Paisley in Scotland for a period of three years. The project aims to develop the technology required in software systems which will be able to provide effective detection and subsequent analysis of fraudulent activity within the general framework required of emerging fixed and mobile telecommunications applications such as electronic and mobile commerce. One postdoctoral position is now available to investigate the application of machine learning and advanced data mining methods in the detection and analysis of anomalous and possibly fraudulent usage of fixed and mobile telecommunications applications such as electronic and mobile commerce. The project will involve the design and implementation of novel algorithms and systems to discover and analyse emerging patterns of anomalous telecommunication system user activity. Highly motivated candidates who have a publication record in, ideally, machine learning, data mining or artificial intelligence applications are encouraged to apply. Applicants should have, or shortly expect to obtain, a PhD in Computer Science. Salaries will be on the R1A scale, starting at 22,200pa to 27,550pa. For further information and informal enquiries please contact Mark Girolami (mark.girolami at paisley.ac.uk). EPSRC & DTI Project Data Mining Tools for Fraud Detection in M-Commerce * DETECTOR http://cis.paisley.ac.uk/giro-ci0/projects.html Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Legal disclaimer -------------------------- The information transmitted is the property of the University of Paisley and is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination and other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer. -------------------------- From terry at salk.edu Tue Apr 2 13:33:59 2002 From: terry at salk.edu (Terry Sejnowski) Date: Tue, 2 Apr 2002 10:33:59 -0800 (PST) Subject: NEURAL COMPUTATION 14:5 Message-ID: <200204021833.g32IXx834531@purkinje.salk.edu> Neural Computation - Contents - Volume 14, Number 5 - May 1, 2002 ARTICLE A Population Study of Integrate-and-Fire-or-Burst Neurons A.R.R. Casti, A. Omurtag, A. Sornborger, E. Kaplan, B. Knight, J. Victor and L. Sirovich NOTE Stable Propagation of Activity Pulses in Populations of Spiking Neurons Werner M. Kistler and Wulfram Gerstner LETTERS Population Coding and Decoding in a Neural Field: A Computational Study Si Wu, Shun-ichi Amari, and Hiroyuki Nakahara The Influence of the Limit Cycle Topology on the Phase Resetting Curve Sorinel A. Oprisan and Carmen C. Canavier Fields As Limit Functions of Stochastic Discrimination and Their Adaptability Philip Van Loocke Learning to Recognize Three-Dimensional Objects Dan Roth, Ming-Hsuan Yang, and Narendra Ahuja A Parallel Mixture of SVMs for Very Large Scale Problems Ronan Collobert, Samy Bengio and Yoshua Bengio Bayesian Framework for Least-Squares Support Vector Machine Classifiers, Gaussian Processes and Kernel Fisher Discriminant Analysis Tony Van Gestel, Johan A.K. Suykens, Gert Lanckriet, Annemie Lambrechts, Bart De Moor and Joos Vandewalle Neural Network Pruning with Tukey-Kramer Multiple Comparison Procedure Donald E. Duckro, Dennis W. Quinn, and Samuel J. Gardner III Products of Gaussians and Probabilistic Minor Component Analysis Chris Williams and Felix Agakov Optimization of the Kernel Functions in a Probabilistic Neural Network Analyzing the Local Pattern Distribution I. Galleske and J. Castellanos Methods for Binary Multidimensional Scaling Douglas L. T. Rohde ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2002 - VOLUME 14 - 12 ISSUES USA Canada* Other Countries Student/Retired $60 $64.20 $108 Individual $88 $94.16 $136 Institution $506 $451.42 $554 * 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 rebollol at aston.ac.uk Wed Apr 3 11:13:41 2002 From: rebollol at aston.ac.uk (laura rebollo-neira) Date: Wed, 03 Apr 2002 16:13:41 +0000 Subject: post doctoral position Message-ID: <3CAB2A35.141E13AB@aston.ac.uk> I would be grateful if you could distribute the following announcement: ASTON UNIVERSITY- SCHOOL OF ENGINEERING & APPLIED SCIENCE RESEARCH ASSOCIATE/FELLOW (3 years fixed term) Starting salary 19.661 pounds Applications are invited for a Post Doctoral Research Associate to work on a 3 years EPSRC funded project entitled "Biorthogonal techniques for optimal signal representation" The project will focus on developing a framework for signal representation outside the traditional orthogonal basis setting, with applications to different areas of signal processing. Further details on the research project can be found in http://www.ncrg.aston.ac.uk/Research.html by clicking on the link for "Biorthogonal techniques for optimal signal representation" Applicants should hold, or expect to hold, a PhD in Physics, Mathematics, Computer Science, Engineering or related discipline. The work requires good mathematical and computational skills and strong interest in some area of signal processing. The successful candidate will have the opportunity to join the the well established and internationally recognized research group NCRG http://www.ncrg.aston.ac.uk The post is tenable from 1 July or as soon as possible thereafter. Informal enquiries are welcomed and should be directed to Dr Laura Rebollo-Neira: rebollol at aston.ac.uk or Prof David Lowe: d.lowe at aston.ac.uk Further particulars and application forms may be obtained electronically on http://www.aston.ac.uk/hr/recruitment.htm or from Personnel Office on Tel (0) 121 359 0870 email b.a.power at aston.ac.uk quoting Ref R02/51 Closing date for applications: 3 May 2002 ~ From gcc at sys.uea.ac.uk Thu Apr 4 07:40:56 2002 From: gcc at sys.uea.ac.uk (Gavin Cawley) Date: Thu, 04 Apr 2002 13:40:56 +0100 Subject: Postdoctoral Research Associate and PhD studentship in Bioinformatics/Machine Learning Message-ID: <5.1.0.14.0.20020404133016.020e38e0@mail.sys.uea.ac.uk> The Exploiting Genomics Programme of the BBSRC and EPSRC has funded a joint project in the John Innes Centre, University of East Anglia and Nottingham University to establish and validate methods to interpret promoter function and establish gene regulatory networks in the model plant Arabidopsis thaliana. A combination of informatics and experimental approaches will be taken using the complete genome sequence, comprising 25,000 annotated genes and extensive full-length cDNA sequence. Dr Gavin Cawley of the School of Information Systems at the University of East Anglia will use bioinformatics and machine learning (including Support Vector Machines) to analyse and classify Arabidopsis promoters using gene expression data from microarrays and gene trap insertion populations. Activities at the UEA require a full time postdoctoral research associate (to be appointed on the Research and Analogous 1A scale, starting salary 20,470 per annum). A further part-time (50%) appointment is available for a postgraduate research associate, who must also register for a PhD. Starting salary will be 9,327 per annum (pro rata to 18,655 per annum) on the Research and Analogous 1B scale. Both posts are for a fixed term of 36 months. Applicants with interests in machine learning (especially kernel methods) and/or bioinformatics are encouraged to apply. Informal enquiries may be made to Dr Cawley by e-mail gcc at sys.uea.ac.uk Applications (stating whether you are applying for the postdoctoral or postgraduate position) in the form of a letter of application, Curriculum Vitae, and the names and addresses of three referees should be sent to Dr Gavin Cawley, School of Information Systems, University of East Anglia, Norwich NR4 7TJ, UK, by no later 19 April 2002. Dr Gavin Cawley School of Information Systems University of East Anglia Norwich NR4 7TJ United Kingdom From becker at nips.salk.edu Mon Apr 8 17:13:46 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:13:46 -0700 (PDT) Subject: NIPS*2002 call for papers Message-ID: __________________________________________________________________ CALL FOR PAPERS Neural Information Processing Systems Natural and Synthetic Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada _________________________________________________________________ Submissions are solicited for the sixteenth meeting of an interdisciplinary conference, which brings together cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians 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. It is single track and highly selective. Preceding the main conference will be one day of tutorial presentations (Dec.9), and following it there will be two days of focused workshops on topical issues at Whistler/Blackcomb ski resort (Dec.13-14). NIPS*2002 INVITED SPEAKERS: Hugh Durrant-Whyte, University of Sydney: Information flow in sensor networks; Paul Glimcher, New York University: Decisions, Uncertainty and the Brain: Neuroeconomics; Deborah Gordon, Stanford University: Ants at Work; Andrew W. Moore, Carnegie Mellon University: Statistical Data Mining; Pietro Perona, Caltech: Learning visual categories; David Heeger, Stanford: Title TBA NIPS*2002 TUTORIAL SPEAKERS: Martin Cooke: Computational auditory scene analysis in listeners and machines; Richard M. Karp: Mathematical, Statistical and Algorithmic Challenges from Genomics and Molecular Biology; Michael Kearns: Computational game theory; Andrew McCallum: Information extraction from the world wide web; Sebastian Seung: Neural integrators; Yair Weiss, Jianbo Shi & Serge Belongie: Eigenvector methods for clustering and image segmentation SUBMISSIONS: The NIPS*2002 categories for paper submission are listed below. The subcategories are by no means exhaustive. Algorithms and Architectures: statistical learning algorithms, neural network architectures, kernel methods, graphical models, Gaussian processes, independent component analysis, model selection, active learning, combinatorial optimization. Applications: innovative applications or fielded systems that use machine learning, including time series, biological applications, text/web analysis, multimedia, robotics, or other intelligent systems. Cognitive Science/Artificial Intelligence: theoretical, computational, or experimental studies of perception, psychophysics, human or animal learning, memory, reasoning, problem solving, language, and neuropsychology. Emerging Technologies: analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, microrobotics, bioMEMS, neural prostheses, photonics, molecular and quantum computing. Neuroscience: theoretical and experimental studies of encoding, decoding, processing, and transmission of information in biological neurons, including spike train generation, synaptic modulation, plasticity and adaptation, and network properties. Reinforcement Learning and Control: Markov decision processes, exploration, planning, navigation, game-playing, multi-agent coordination, computational models of classical and operant conditioning. Speech and Signal Processing: recognition, coding, synthesis, de-noising, source separation, auditory perception, psychoacoustics, temporal algorithms for signal processing such as Markov models, dynamical systems, recurrent networks. Theory: learning theory, information theory, statistical physics of learning, Bayesian methods, approximation bounds, online learning and dynamics, generalization and regularization. Visual Processing: 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 newly created demonstration track should consult the Web site below for more detailed instructions. REVIEW CRITERIA: All submitted papers will be thoroughly refereed on the basis of technical quality, significance, and clarity. Authors new to NIPS are particularly encouraged to submit their work. There will be an opportunity after the meeting to revise accepted manuscripts before submitting a final camera-ready copy for the proceedings. PAPER FORMAT: Submitted papers 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 8.25in by 5in rectangle. Submissions failing to follow these guidelines will not be considered. Authors are strongly encouraged to use the NIPS LaTeX style files obtainable from the web site listed below. SUBMISSION INSTRUCTIONS: NIPS accepts only electronic submissions. Full submission instructions will be available at the web site given below. NIPS accepts submissions in postscript and PDF format. The electronic submission process will begin on June 9, 2002. SUBMISSION DEADLINE: SUBMISSIONS MUST BE LOGGED BY MIDNIGHT JULY 1, 2002, PACIFIC DAYLIGHT TIME. The LaTeX style files, the electronic submission page, and other conference information are available on the web at http://nips.cc NIPS*2002 ORGANIZING COMMITTEE: General Chair, Sue Becker, McMaster University; Program Chair, Sebastian Thrun, Carnegie Mellon University; Publications Chair, Klaus Obermayer, TU Berlin; Tutorial Chair, Lawrence Saul, University of Pennsylvania; Workshops Co-chairs, Barak Pearlmutter, University of New Mexico, Robert Jacobs, University of Rochester; Demonstrations Co-chairs, Shihab Shamma, University of Maryland, Shih-Chii Liu, University and ETH Zurich; Publicity Chair, Zoubin Ghahramani, University College London; Volunteer Coordinator, Rajesh Rao, University of Washington; Treasurer, Bartlett Mel, University of Southern California; Web Masters, Alex Gray, Carnegie Mellon University, Guy Lebanon, Carnegie Mellon University; Contracts, Steve Hanson, Rutgers University. NIPS*2002 PROGRAM COMMITTEE: Sebastian Thrun, Carnegie Mellon University (chair); Peter Bartlett, BIOwulf Technologies and University of California, Berkeley; Gert Cauwenberghs, Johns Hopkins University; Geoffrey Gordon, Carnegie Mellon University; Daniel Lee, University of Pennsylvania; Marina Meila, University of Washington; Klaus-Robert Mueller, Fraunhofer FIRST ; Andrew Y. Ng, University of California, Berkeley; John Platt, Microsoft Research; Sam Roweis, University of Toronto; Eero Simoncelli, New York University; Joshua Tenenbaum, MIT; Chris Williams, University of Edinburgh; Richard Zemel, University of Toronto. PAPERS MUST BE SUBMITTED BY JULY 1, 2002 -- PLEASE POST -- From becker at nips.salk.edu Mon Apr 8 17:14:48 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:14:48 -0700 (PDT) Subject: NIPS*2002 call for workshop proposals Message-ID: _________________________________________________________________ Neural Information Processing Systems Natural and Synthetic NIPS*2002 Post-Conference Workshops December 13 and 14, 2002 Call for Workshop Proposals Whistler/Blackcomb Resort, BC, CANADA _________________________________________________________________ Following the regular program of the Neural Information Processing Systems 2002 conference in Vancouver, BC, Canada, workshops on various current topics in neural information processing will be held on December 13 and 14, 2002, in Whistler, BC, Canada. We invite researchers interested in chairing one of these workshops to submit workshop proposals. 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 particularly encouraged. Workshop topics include, but are not limited to, the following: Active Learning, Attention, Audition, Bayesian Analysis, Bayesian Networks, Benchmarking, Brain Imaging, Computational Complexity, Computational Molecular Biology, Control, Genetic Algorithms, Graphical Models, Hippocampus and Memory, Hybrid Supervised/Unsupervised Learning, Hybrid HMM/ANN Systems, Implementations, Independent Component Analysis, Mean-Field Methods, Markov Chain Monte-Carlo Methods, Music, Network Dynamics, Neural Coding, Neural Plasticity, On-Line Learning, Optimization, Recurrent Nets, Robot Learning, Rule Extraction, Self-Organization, Sensory Biophysics, Signal Processing, Spike Timing, Support Vectors, Speech, Time Series, Topological Maps, and Vision. Detailed descriptions of previous workshops may be found at http://www-2.cs.cmu.edu/Web/Groups/NIPS/NIPS2001/prevconf.html. There will be six hours of workshop meetings per day, split into morning and afternoon sessions, with free time in between 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 evening plenary sessions -- Writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. SUBMISSION INSTRUCTIONS Interested parties should submit a short proposal for a workshop of interest via email by August 9, 2002. Proposals should include 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 pure "mini-conference" format. An example format is: Tutorial lecture providing background and introducing terminology relevant to the topic. Two short lectures introducing different approaches, alternating with discussions after each lecture. 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% 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 motivate why the topic is of interest or controversial, why it should be discussed, and who the targeted group of participants is. It also should 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-workshop-proposal at cs.unm.edu. Please do not use attachments, Microsoft Word, postscript, html, or pdf files. Questions may be addressed to nips-workshop-admin at cs.unm.edu. Information about the main conference and the workshop program can be found at http://nips.cc/. Barak A. Pearlmutter, University of New Mexico Robert A. Jacobs, University of Rochester NIPS*2002 Workshops Co-Chairs PROPOSALS MUST BE RECEIVED BY AUGUST 9, 2002 --- Please Post --- From becker at nips.salk.edu Mon Apr 8 17:15:41 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:15:41 -0700 (PDT) Subject: NIPS*2002 call for demonstrations Message-ID: *** New at NIPS: Demonstrations Track *** __________________________________________________________________ Neural Information Processing Systems Natural andSynthetic CALL FOR DEMONSTRATIONS Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada} __________________________________________________________________ For the first time, the Neural Information Processing Systems conference will include a separate track for demonstrations. The demonstrations will take place in parallel with the poster sessions at the NIPS*2002 conference. Example areas of interest for the demonstrations track include but are by no means limited to the following: Analog and digital VLSI Neuromorphic Engineering Computational sensors and actuators Robotics bioMEMS (microelectromechanical systems) Biomedical instrumentation Neural prostheses Photonics Real-time multimedia systems Large-scale neural emulators Software demonstrations of novel algorithms NIPS is an interdisciplinary conference, which attracts cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The demonstration track enables researchers to highlight scientific advances, systems, and technologies in ways that go beyond conventional poster presentations. It will provide a unique forum for demonstrating advanced technologies (hardware and software), and fostering the direct exchange of knowledge. We hope that this track will stimulate interactions between researchers from different fields (for example, roboticists and neuromorphic engineers) and encourage new colloboration between researchers in theoretical fields and those in more applied fields. Submissions accepted in the demonstrations track will be published on the NIPS web site, but will not appear in printed proceedings. However, submitting your work to the demonstration track by no means precludes the submission of a companion paper to the regular NIPS conference. In fact, joint submissions are very much encouraged. We also encourage authors submitting demonstrations to consider organizing a workshop at NIPS*2002. Note that the deadline for paper submissions is July 1, 2002, and for demonstration and workshop proposals the submission deadline is August 9, 2002. Please see http://nips.cc for further details. There will be a separate room for these demonstrations and participants will have access to power strips, tables and poster boards. VCRs and monitors will also be provided on request. Participants are responsible for ensuring that their demonstration is sufficiently portable; additional hardware beyond that specified above will not be provided by NIPS. SUBMISSION INSTRUCTIONS: All proposals for demonstrations will be reviewed by the Demonstrations Co-Chairs. Interested parties should submit a brief description of their proposed demonstration via email by August 9, 2002. Proposals should include a title, description of the device or system to be demonstrated, main results, novelty and significance of the work, any related publications, and estimated space requirements for the demonstration. Please include the name, address, email address, phone and fax numbers for all co-authors on the submmitted work, and indicate whether a related paper has also been submitted to NIPS*2002. Proposals should be emailed to shih at ini.phys.ethz.ch, and should be in plain ascii text, postscript or pdf. Questions may be addressed to shih at ini.phys.ethz.ch. Information about the main conference and the workshop program can be found at http://nips.cc Shihab Shamma Electrical Engineering Department & Institute for Systems Research University of Maryland and Shih-Chii Liu Institute of Neuroinformatics, University of Zurich & ETH Zurich NIPS*2002 Demonstrations Co-Chairs PROPOSALS MUST BE RECEIVED BY AUGUST 9, 2002 --- Please Post --- From rsun at cecs.missouri.edu Mon Apr 8 17:12:33 2002 From: rsun at cecs.missouri.edu (rsun@cecs.missouri.edu) Date: Mon, 8 Apr 2002 16:12:33 -0500 Subject: Cognitive Systems Research, Volume 3, Issue 1 Message-ID: <200204082112.g38LCXB08496@ari1.cecs.missouri.edu> A new issue is now available for * Cognitive Systems Research, Volume 3, Issue 1, Pages 1-119 (March 2002) =============================================================================== Cognitive Systems Research, Volume 3, Issue 1, Pages 1-119 (March 2002) TABLE OF CONTENTS Introduction to the special issue on computational cognitive modeling, Pages 1-3 Christian D. Schunn and Wayne D. Gray http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-3/1/711b118ed203aeda9a398ad63954e8dc A modular neural-network model of the basal ganglia's role in learning and selecting motor behaviours, Pages 5-13 Gianluca Baldassarre http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-1/1/35c9176c4afb2fb3a3ba329b618af22a Bootstrapping in miniature language acquisition, Pages 15-23 Rutvik Desai http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-4/1/747206abfb974b02bf41c156c86950bb Modeling icon search in ACT-R/PM, Pages 25-33 Michael D. Fleetwood and Michael D. Byrne http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-5/1/aa433104bb8019311d4f5630752c4b8a In search of templates, Pages 35-44 Fernand Gobet and Samuel Jackson http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-6/1/b8468b413f4ca0b9c2471e277a46095c An attractor network model of serial recall, Pages 45-55 Matt Jones and Thad A. Polk http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-7/1/b31a07d56e1992bf0131d5ffd6b0cf95 Intention superiority effect: A context-switching account, Pages 57-65 Christian Lebiere and Frank J. Lee http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-8/1/9c02678c6285705623482bff50cb8ca1 Modeling selective attention: Not just another model of Stroop (NJAMOS), Pages 67-76 Marsha C. Lovett http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-2/1/416e6f6b5973838ca291ccf14a772ec6 Extending task analytic models of graph-based reasoning: A cognitive model of problem solving with Cartesian graphs in ACT-R/PM, Pages 77-86 David Peebles and Peter C. -H. Cheng http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-3/1/c0b76a0b1f47140f3067fddc1079731d The role of computational modeling in understanding hemispheric interactions and specialization, Pages 87-94 James A. Reggia and Reiner Schulz http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-1/1/319c267472df4cc3919c2103d3d425ba Predicting the effects of cellular-phone dialing on driver performance, Pages 95-102 Dario D. Salvucci and Kristen L. Macuga http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-9/1/57923a5f59840882407bfd66fb61a599 A model of individual differences in skill acquisition in the Kanfer-Ackerman air traffic control task, Pages 103-112 Niels A. Taatgen http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-2/1/f2a7daf9433c6074808d7fd946c19afc An explanation of the length effect for rotated words, Pages 113-119 Carol Whitney http://www.sciencedirect.com/science/article/B6W6C-44HTMB3-1/1/452c70ac314fb5e174c1a994a7103278 =============================================================================== See the following journal Web pages for subscription information: http://www.elsevier.com/locate/cogsys http://www.cecs.missouri.edu/~rsun/journal.html =========================================================================== Prof. Ron Sun http://www.cecs.missouri.edu/~rsun CECS Department phone: (573) 884-7662 University of Missouri-Columbia fax: (573) 882 8318 201 Engineering Building West Columbia, MO 65211-2060 email: rsun at cecs.missouri.edu http://www.cecs.missouri.edu/~rsun http://www.cecs.missouri.edu/~rsun/journal.html http://www.elsevier.com/locate/cogsys =========================================================================== From J.A.Bullinaria at cs.bham.ac.uk Tue Apr 9 08:01:02 2002 From: J.A.Bullinaria at cs.bham.ac.uk (John A Bullinaria) Date: Tue, 9 Apr 2002 13:01:02 +0100 (BST) Subject: Lectureships / Seinor Lectureships at Birmingham, UK Message-ID: Dear Connectionists, The University of Birmingham is looking for three additional Lecturers / Seinor Lecturers in Computer Science, Software Engineering or Artificial Intelligence. The specific areas of research are open. Birmingham already has a strong research group in natural computation, and we would warmly encourage outstanding applicants from the broad field of nature inspired computation (including neural networks, evolutionary computation, etc.). There are currently eight members of academic staff working on natural computation and related topics: Dr. John Bullinaria (Neural Networks, Evolutionary Computation, Cog.Sci.) Dr. Ke Chen (Neural Networks, Pattern Recognition, Machine Perception) Dr. Aniko Ekart (Genetic Programming, AI, Machine Learning) Dr. Jun He (Evolutionary Computation, Artificial Immune Systems) Dr. Julian Miller (Evolutionary Computation, Machine Learning) Dr. Jon Rowe (Evolutionary Computation, AI) Dr. Thorsten Schnier (Evolutionary Computation, Engineering Design) Prof. Xin Yao (Evolutionary Computation, NNs, Nature Inspired Comp.) Other staff members also working in these areas include Prof. Aaron Sloman (evolvable architectures of mind, co-evolution, interacting niches), Dr. Jeremy Wyatt (evolutionary robotics, reinforcement learning), and Dr Ela Claridge (evolutionary image processing). The official job advert is attached below. Best regards, John Bullinaria. ========================================================================== THE UNIVERSITY OF BIRMINGHAM School of Computer Science (http://www.cs.bham.ac.uk/) NEW ACADEMIC POSTS AVAIABLE IN COMPUTER SCIENCE, SOFTWARE ENGINEERING OR ARTIFICIAL INTELLIGENCE. Applications are invited for three open-ended (i.e. not temporary) appointments, at the level of Lecturer or Senior Lecturer, in Computer Science, Software Engineering or Artificial Intelligence. One post is available to start from 1 September 2002 (or later), the remaining two will be available from 01 January 2003 (or later): exact start dates are negotiable. The appointees will be expected to contribute stongly to research, teaching and administration within the School. We would welcome applicants who wish to lead the School into new research or teaching areas as well as those who wish to add to the School's existing strengths. The School also encourages industrial collaboration and consultancy. The School obtained a grade of 5 in the 2001 national Research Assessment Exercise. This shows international recognition of the excellence of the School's research. The School is now starting a new phase of expansion over several years which is planned to include the appointment of a new professor (equivalent to full professor in the USA). Several additional posts will be advertised later, with start dates in 2003. Applicants should have an internationally excellent research background in Computer Science, Software Engineering or Artificial Intelligence, as evidenced by publications in leading international journals or conference proceedings in these fields. Applicants should have teaching experience in one of the fields, and should have, or expect soon to have, a PhD in one of the fields or an appropriate, closely related discipline. Applicants for a Senior Lecturership should already meet our criteria for promotion to that level. For these criteria, please contact Mr David King, tel: (+44)(0)121 414 3711; Email: D.J.King at cs.bham.ac.uk. Starting salary for Lecturers on scale £20,470 - £32,537 a year depending on experience and qualifications. Starting salary for Senior Lecturers on scale £34,158 - £38,603 year depending on experience and qualifications. (The university pays additional pension contributions.) Informal enquiries to: Professor Aaron Sloman Tel: (+44) (0)121 414 3711 or email: A.Sloman at cs.bham.ac.uk If possible please avoid sending him Word files: use plain text or html, or postscript or PDF. Please state in any application whether you are interested in a Lectureship or Senior Lectureship. Application forms returnable by 25 April 2002 (late applications may be considered). Further particulars including instructions on how to apply, and pointers to online application forms, can be found at: http://www.cs.bham.ac.uk/news/jobs/Lectureship2002.html Please quote reference S35803/02. Further information about the School and its Research and teaching can be found here: http://www.cs.bham.ac.uk/ http://www.cs.bham.ac.uk/research/ http://www.cs.bham.ac.uk/news/ http://www.cs.bham.ac.uk/study/ Working towards equal opportunities. ========================================================================== From ken at phy.ucsf.edu Tue Apr 9 18:57:05 2002 From: ken at phy.ucsf.edu (Ken Miller) Date: Tue, 9 Apr 2002 15:57:05 -0700 Subject: Paper available: reverse suture and orientation map development Message-ID: <15539.29121.257475.147355@coltrane.ucsf.edu> The following paper is now available from ftp://ftp.keck.ucsf.edu/pub/ken/miller-erwin01.ps (postscript, 4 MB) or ftp://ftp.keck.ucsf.edu/pub/ken/miller-erwin01.ps.gz (gzipped ps, 1.2 MB) or from http://www.keck.ucsf.edu/~ken (click on 'publications', then on 'models of neural development') "Effects of Monocular Deprivation and Reverse Suture On Orientation Maps Can Be Explained By Activity-Instructed Development of Geniculocortical Connections" by K.D. Miller and E. Erwin This is a final draft of a paper that has now appeared as Visual Neuroscience 18:821-834 (2001). Abstract: Mature visual cortex shows a single, binocularly matched orientation map. This matching develops without visual experience. It persists despite early monocular deprivation that largely eliminates one eye's map, followed by reverse suture (deprivation of the previously open eye and opening of the previously deprived eye), even though the two eyes lack common visual experience in this case. These results have been interpreted to suggest that the structure of orientation maps either is innately predetermined or, if it arises through self-organization, is determined by external cues such as boundary conditions or a ``scaffolding'' of horizontal connections. We show, to the contrary, that these results are the expected outcomes if orientation maps develop through activity-instructed, correlation-based development of the geniculocortical connections without additional cues. A weak, binocularly correlated orientation map is known to exist before deprivation onset; we previously showed how this can arise through activity-instructed development. Now we show that this initial correlation between the two eyes' maps can persist or increase despite deprivation sufficient to cause massive loss of the deprived eye's geniculocortical synaptic strength, followed by reverse suture. Given sufficient early correlated map development, each map's fate is ``dynamically committed'': the two eyes' maps will converge upon a common outcome, even if developing independently. This dynamic fate commitment is retained even after severe deprivation. Ken Kenneth D. Miller telephone: (415) 476-8217 Associate Professor fax: (415) 476-4929 Dept. of Physiology, UCSF internet: ken at phy.ucsf.edu 513 Parnassus www: http://www.keck.ucsf.edu/~ken San Francisco, CA 94143-0444 From castellano at di.uniba.it Thu Apr 11 06:37:59 2002 From: castellano at di.uniba.it (Giovanna Castellano) Date: Thu, 11 Apr 2002 12:37:59 +0200 Subject: call for contribution at invited session in KES2002 Message-ID: <00af01c1e144$f31eeb00$c9bbccc1@castellano> As you may aware, the coming KES Conference will be held in September this year in Italy (http://www.bton.ac.uk/kes/kes2002/) I am currently organizing a special session for the conference in the area of Knowledge-based neurocomputing systems. I would like to invite you to submit papers to this session. The original deadline for submissions is 1 April. However, with the understanding that many of us may be busy and unable to meet this deadline, I have negotiated a further extension of the deadline to 20 April. So if you are interested in contributing to this session please let me know at your earliest convenience, sending ASAP your intention to submit a paper to me (castellano at di.uniba.it)l. You can find below the cfp containing information for paper submission. If you have any questions in that regard please feel free to contact me. Dr. Giovanna CASTELLANO Assistant Professor, PhD in Computer Science Computer Science Department University of Bari Via Orabona, 4 70125 Bari, ITALY phone: +39 080 5442456 fax : +39 080 5443196 e-mail: castellano at di.uniba.it URL: www.di.uniba.it/people/castellano.htm ================================================================ KES'2002 Sixth International Conference on Knowledge-Based Intelligent Information Engineering Systems 16, 17 & 18 September 2002 Podere d'Ombriano, Crema, Italy Special Session: Knowledge-Based Neurocomputing Systems Call for papers This special session aims at presenting and discussing Knowledge-Based Neurocomputing Systems, which concern the use and the explicit representation of problem-specific knowledge within the neurocomputing paradigm. All areas of simulations where KBN systems can be applied will be considered, giving a special focus to applications in medicine. Topics of interest include (but are not limited to): a.. Hybrid architectures for neural networks b.. Fuzzy and neurofuzzy techniques c.. Methods of rule extraction/discovery d.. Clustering and classification e.. Pattern recognition f.. Predictive models g.. Control / Optimization systems h.. Medical and diagnostic systems Submissions Papers should be submitted in electronic format (preferably in PDF or PostScript) to castellano at di.uniba.it, using the subject-line "KES02 Special Session submission". Papers must be formatted as detailed in IOS Instructions for the Preparation of a Camera-Ready Manuscript (see the KES 2002 homepage for details). The maximum length of papers is five pages (when prepared according to the IOS instructions). Longer papers will be subject to an additional page charge. Please note that final versions of accepted papers must be submitted in hard copy. Submitted papers have to be unpublished, containing new and original results. Submission implies the willingness of at least one of the authors to register and present the paper. The session papers will be published in the Conference Proceedings (IOS Press, Amsterdam, the Netherlands). Extended versions of selected papers will be considered for publication in the KES Journal (International Journal of Knowledge-Based Intelligent Engineering Systems). Important Dates Submission deadline: April 20, 2002 Notification of acceptance: May 15, 2002 Deadline for camera-ready papers: June 1, 2002 Session Organizer and Chair Dr. Giovanna Castellano Computer Science Department University of Bari Via Orabona, 4 70125 Bari, ITALY Tel: +39 080 5442456 Fax: +39 080 5443196 email: castellano at di.uniba.it From shastri at ICSI.Berkeley.EDU Thu Apr 11 15:04:19 2002 From: shastri at ICSI.Berkeley.EDU (Lokendra Shastri) Date: Thu, 11 Apr 2002 12:04:19 PDT Subject: Paper on computational abstraction of long-term potentiation Message-ID: <200204111904.MAA22112@dill.ICSI.Berkeley.EDU> Dear Connectionists, The following article may be of interest to you: "A Computationally Efficient Abstraction of Long-term Potentiation" Neurocomputing (In Press). The final draft is available at: http://www.icsi.berkeley.edu/~shastri/psfiles/ShastriNCltp02.pdf Best wishes, -- Lokendra Shastri http://www.icsi.berkeley.edu/~shastri Phone: (510) 666-2910 FAX: (510) 666-2956 --------------------------------------------------------------------- A Computationally Efficient Abstraction of Long-term Potentiation Lokendra Shastri ICSI, Berkeley, CA. Key words: Long-term potentiation; hippocampus; binding detection; dentate gyrus; episodic memory Abstract A computational abstraction of long-term potentiation (LTP) is proposed. The abstraction captures key temporal and cooperative properties of LTP, and also lends itself to rapid computation. The abstraction is used to simulate the recruitment of binding-detector cells in the dentate gyrus (DG). The simulation results are used to validate a quantitative analysis of binding-detector cell recruitment in DG. The analysis shows that (i) a large number of binding-detector cells would be recruited in response to entorhinal cortex activity and (ii) cells recruited for distinct bindings would exhibit very low cross-talk. These results help in understanding the neural basis of episodic memory. From tbl at cin.ufpe.br Fri Apr 12 10:21:25 2002 From: tbl at cin.ufpe.br (Teresa Bernarda Ludermir) Date: Fri, 12 Apr 2002 11:21:25 -0300 Subject: Deadline Extension - VII BraziDeadline Extension - VII BrazilianDeadline Extension - VII Brazilian Symposium on Neural Networks (SBRN 2002) Message-ID: <3CB6ED65.C12EBAC7@cin.ufpe.br> Because of numerous requests for extensions, the deadline for submission of papers to the VII Brazilian Symposium on Neural Networks (SBRN 2002) is now April 22, 2002. This symposium will be held in Porto de Galinhas, Recife, Brazil from November 11-14, 2002, and is chaired by Professor Teresa Ludermir from the Federal University of Pernambuco in Brazil. All information regarding this symposium can be found on our website at http://www.cin.ufpe.br/~sbiarn02/sbrn02.html Please refer to this website for formatting and submission requirements. If you have any questions about the symposium, please feel free to contact us at sbrn2002 at cin.ufpe.br. Sincerely, Marcilio C. P. de Souto Program Chair - SBRN 2002 -- Teresa B. Ludermir http://www.cin.ufpe.br/~tbl Centro de Informatica - UFPE Phone: [+55 81] 3271-8430 Caixa Postal 7851 x: 4308 CEP 50732-970 Recife PE Brazil From sfr at unipg.it Fri Apr 12 09:11:58 2002 From: sfr at unipg.it (Simone G.O. Fiori (Pg)) Date: Fri, 12 Apr 2002 15:11:58 +0200 (DFT) Subject: Announcement of Tutorial on Neural Networks. Message-ID: <1018617118.3cb6dd1e8b8bb@webmail.unipg.it> Dear Colleagues, I am pleased to announce that during the Tenth Biennial IEEE Conference on Electromagnetic Field Computation (CEFC), to be held in Perugia (Italy) on June 15-19, 2002, a tutorial about the application of artificial neural networks to electromagnetic compatibility will be given. The tutorial is intended for young investigators interested in the roots of neural networks theory as well as for senior researchers in the neural field interested in the latest issues about networks' application to electro- magnetics. The tutorial program includes therefore topics such as the basic theory of multilayer perceptrons and their applications as well as emerging neural techniques such as independent component analysis. The title and time-scheduling of the full-day tutorial are: Neural Network Modeling of Electromagnetic Interaction Phenomena (Instructor: Dr. Simone Fiori, Perugia University) Saturday, June 15, 2002 (Time-scheduling: 9,00-13,00 and 15,00-18,00) The cost* for the attendances is: EUR 100/full registration including course notes, EUR 65/student registration including course notes. * Please note that, in order to attend the tutorial, it is not mandatory to subscribe to the conference. A short description of the conference is attached below for your convenience. For more information please visit the conference web-page at: http://www.unipg.it/cefc2002/ or contact the instructor (S. Fiori) at the email address sfr at unipg.it or the conference general chair (E. Cardelli) at the email address cefc2002 at unipg.it. Sincerely, S. Fiori Faculty of Engineering, Perugia University (Italy) ================================================================== The Tenth Biennial IEEE Conf. on Electromagnetic Field Computation: The Tenth Biennial IEEE Conference on Electromagnetic Field Computation (CEFC) will be held at Perugia during June 15-19, 2002. Perugia is an ancient town founded in the prehistoric epoch, located in the Umbria region, also called "the green heart of Italy" for its rich and extensive vegetation. Perugia is a very popular tourist destination, known for its important monuments, its folklore, its rich cuisine and its quiet and friendly people. The last Conference was held in Milwaukee, Wisconsin, USA in 2000. The aims of the IEEE CEFC are to present the latest developments in modelling and simulation methodologies for the analysis of electromagnetic fields and wave interactions, with the application emphasis being on the computer-aided design of low and high frequency devices, components and systems. Scientists and engineers worldwide will present and discuss original contributions in the areas of static and quasi-static fields; wave propagation; material modelling; coupled problems, optimisation; numerical techniques, software methodology; applications of electromagnetic CAD to electrical/electronic devices, components and systems prototyping. The Conference will feature oral and poster presentations. From l.s.smith at cs.stir.ac.uk Fri Apr 12 08:16:33 2002 From: l.s.smith at cs.stir.ac.uk (Leslie Smith) Date: Fri, 12 Apr 2002 13:16:33 +0100 Subject: 2 research assistant posts at Stirling University (Scotland) Message-ID: <3CB6D021.BD8FDA05@cs.stir.ac.uk> University of Stirling Department of Computing Science and Mathematics Research Assistant (2 posts) UK?17,626 - UK?19,681 The University of Stirling (jointly with the Universities of Glasgow and Oxford) has recently been awarded two major EPSRC research grants on communication between real neurons and electronics, and on spiking neural systems. Consequently, it has the following opportunities: Post 1: Communication between real neurons and electronics: Signalling between cultured neurons and electronics Three year Research Assistant post concerned primarily with the development of signals to be applied to neural systems, and with the interpretation of spikes emitted by these neural systems. Reference No: 6156/920 Post 2: Spiking neural systems: Spiking in low-level auditory processing Three year Research Assistant post developing neuromorphic spike-based algorithms for implementation in silicon (at Oxford and Edinburgh) for solving problems in early auditory processing. Reference No: 6157/927 For informal discussion contact Prof. Leslie Smith, lss at cs.stir.ac.uk tel (44) 1786 467435. For more information about the Department please see www.cs.stir.ac.uk. There will also be related posts at Edinburgh: see www.ee.ed.ac.uk/ISG/ Further particulars are available from the Personnel Office, University of Stirling, Stirling, FK9 4LA. Tel: (01786) 467028, Fax: (01786) 466155, e-mail personnel at stir.ac.uk Closing date for applications: Monday 6 May 2002. www.personnel at stir.ac.uk AN EQUAL OPPORTUNITIES EMPLOYER -- Professor Leslie S. Smith, Head of Department, Dept of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA, Scotland l.s.smith at cs.stir.ac.uk Tel (44) 1786 467435 Fax (44) 1786 464551 www http://www.cs.stir.ac.uk/~lss/ From D.Willshaw at anc.ed.ac.uk Mon Apr 15 07:05:15 2002 From: D.Willshaw at anc.ed.ac.uk (David Willshaw) Date: Mon, 15 Apr 2002 12:05:15 +0100 (BST) Subject: Edinburgh Doctoral Training in Neuroinformatics Message-ID: <15546.46059.403874.353053@gargle.gargle.HOWL> ------------------------------------------------------------------------ 4 YEAR DOCTORAL TRAINING IN NEUROINFORMATICS http://www.anc.ed.ac.uk/neuroinformatics first application deadline: May 1st The University of Edinburgh has just established a Doctoral Training Centre in Neuroinformatics. The programme is made up of 3 themes: 1) Computational and Cognitive Neurocience - analytical and computational modelling of information-processing in the nervous system; 2) Neuromorphic Engineering and Neural Computation - Artificial sensor perception and analysis, neuromorphic modelling, spiking computation and mixed-mode VLSI, evolutionary optimisation of physical systems, neurally inspired algorithms; 3) Simulation, Analysis, Visualisation and Data Handling - software systems and computational techniques for neuroscience and neural engineering The 4 year programme in Neuroinformatics consists of an introductory year of training in neuroscience and this broad range of application areas culminating in an MSc by Research in Neuroinformatics. This is followed by 3 years of Ph.D. study, building on the research project undertaken in the first year. Applications for 4 year studentships for entry in October 2002 are invited from EU students (see website for full eligibility). The stipend is set in the region of 13,000 pounds per annum. Full fees are paid. Note particularly that this studentship scheme is open to all students within the EU and is not restricted to applicants from the UK. Students with a strong background in computer science, electronics, mathematics, physics or engineering are particularly welcome, and we will consider those with other backgrounds. We will be accepting an average of 10 students a year into this programme, which will be located within the Institute for Adaptive and Neural Computation in the Division of Informatics, the UK's largest highest-quality academic computer-science grouping. Additional faculty are being recruited to support the Doctoral Training Centre and to grow Neuroinformatics at Edinburgh. Edinburgh has strong programmes of research in all of the areas listed above and leads the UK in integrating these into a coherent programme in Neuroinformatics. For full application details, and further information of areas of study at the Centre, consult the website: http://www.anc.ed.ac.uk/neuroinformatics or send email to neuroinformatics-phd at anc.ed.ac.uk Applications received by MAY 1 2002 will receive priority treatment. From bengio at idiap.ch Mon Apr 15 08:21:16 2002 From: bengio at idiap.ch (Samy Bengio) Date: Mon, 15 Apr 2002 14:21:16 +0200 (CEST) Subject: open postdoc positions in Machine Learning in Switzerland Message-ID: Open positions for Postdoctoral candidates in Machine Learning The Dalle Molle Institute for Perceptual Artificial Intelligence (IDIAP, http://www.idiap.ch) seeks qualified applicants for Postdoc positions in its Machine Learning group for research projects related to multimodal integration of asynchronous streams of information. The ideal candidate should have strong background in statistical learning theory, mixture models, and more specifically hidden Markov models and related sequence modeling tools. All applicants should be familiar with C/C++ programming under a Unix environment. Although IDIAP is located in the French part of Switzerland, English is the main working language at IDIAP. Free English and French courses are also provided. IDIAP is located in the town of Martigny (http://www.martigny.ch) in Valais, a scenic region in the south of Switzerland, surrounded by the highest mountains of Europe, and offering exciting recreational activities, including hiking, climbing and skiing, as well as varied cultural activities. It is within close proximity to Montreux (Jazz Festival) and Lausanne (EPFL, http://www.epfl.ch). Candidates should send their detailed CV to Dr. Samy Bengio Research Director. Machine Learning Group Leader. IDIAP, CP 592, rue du Simplon 4, 1920 Martigny, Switzerland. tel: +41 27 721 77 39, fax: +41 27 721 77 12. mailto:bengio at idiap.ch, http://www.idiap.ch/~bengio ----- Samy Bengio Research Director. Machine Learning Group Leader. IDIAP, CP 592, rue du Simplon 4, 1920 Martigny, Switzerland. tel: +41 27 721 77 39, fax: +41 27 721 77 12. mailto:bengio at idiap.ch, http://www.idiap.ch/~bengio From hamilton at may.ie Mon Apr 15 17:20:26 2002 From: hamilton at may.ie (Hamilton Institute) Date: Mon, 15 Apr 2002 22:20:26 +0100 Subject: Statisical Machine Learning PhD Position - Hamilton Institute/DaimlerChrysler Research Message-ID: <007b01c1e4c3$5dc7b7d0$06c09d95@pwhm6> JOINT PHD POSITION, HAMILTON INSTITUTE/DAIMLERCHRYSLER RESEARCH The Hamilton Institute invites applications for a PhD position in the general area of advanced automotive control. The project will involve utilising and developing concepts in some or all of the following areas: decentralised and interacting control systems; hybrid dynamical systems; statistical machine learning. This post is part of a collaboration with DiamlerChrysler Research (Stuttgart), the Center for Systems Science (Yale University) and the Hamilton Institute (NUI Maynooth). Candidates will be primarily based at the Hamilton Institute and at DaimlerChrysler Research. Successful candidates are likely to have demonstrated an outstanding level of academic achievement. These posts offer an exciting opportunity to tackle fundamental research problems within a stimulating multi-disciplinary research environment with state of the art facilities and strong links to the international research community. Applicants should send a C.V., including details of three referees to hamilton at may.ie. For further details, please visit www.hamilton.may.ie. The closing date for applications is the 31st May 2002. From Zoubin at gatsby.ucl.ac.uk Wed Apr 17 20:43:13 2002 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 18 Apr 2002 01:43:13 +0100 (BST) Subject: NIPS 2001 Proceedings Now Available Online Message-ID: <200204180043.BAA20546@crick.gatsby.ucl.ac.uk> The full proceedings of the 2001 Neural Information Processing Systems Conference are now available online at: http://www.nips.cc/2001papers These will be published in Advances in Neural Information Processing Systems 14. T. G. Dietterich, S. Becker, and Z. Ghahramani (eds.) MIT Press, Cambridge, MA, 2002. Please note that the FULL TEXT of the proceedings is searchable if you have the djvu-plugin. We hope you enjoy this resource. Zoubin Ghahramani NIPS 2001 Publications Chair From wahba at stat.wisc.edu Wed Apr 17 17:52:16 2002 From: wahba at stat.wisc.edu (Grace Wahba) Date: Wed, 17 Apr 2002 16:52:16 -0500 (CDT) Subject: The Multicategory Support Vector Machine Message-ID: <200204172152.QAA03061@juno.stat.wisc.edu> I'm pleased to announce "Classification of Multiple Cancer Types by Multicategory Support Vector Machines Using Gene Expression Data" by Yoonkyung Lee and Cheol-Koo Lee, UW-Madison Statistics Dept TR 1051 (2002). This paper (and related papers) are accessible via the home pages: http://www.stat.wisc.edu/~yklee or http://www.stat.wisc.edu/~wahba Abstract: Monitoring gene expression profiles is a novel approach in cancer diagnosis. Several studies showed that prediction of cancer types using gene expression data is promising and very informative. The Support Vector Machine (SVM) is one of the classification methods successfully applied to the cancer diagnosis problems using gene expression data. However, its extension to more than two classes was not obvious, which might impose limitations in its application to multiple tumor types. In this paper, we analyze a couple of published multiple cancer types data sets by the multicategory SVM, which is a recently proposed extension of the binary SVM. ...................... ....................... The Multicategory Support Vector Machine was proposed in "Multicategory Support Vector Machines", Yoonkyung Lee, Yi Lin and Grace Wahba UW-Madison Statistics Dept TR 1043 (2001), also accessible as above. ......................... Multicategory `Soft' classification was proposed in "Smoothing Spline Analysis of Variance for Polychotomous Response Data" Xiwu Lin, UW-Madison Statistics Dept TR 1003 (1998), available via the wahba home pg. It can be argued that the SVM can be considered a `hard' classifier and while the penalized likelihood estimate (as in TR 1003)is a `soft' classifier, and it can be argued that the the SVM is more appropriate where the attribute data is relatively sparse and the category overlap is relatively small while the soft classifier may be more appropriate where the attribute data is relatively dense and the category overlap is more substantial. From ole.jensen at fcdonders.kun.nl Wed Apr 17 08:40:04 2002 From: ole.jensen at fcdonders.kun.nl (Ole Jensen) Date: Wed, 17 Apr 2002 14:40:04 +0200 Subject: paper announcement : evoked gamma Message-ID: <3CBD6D24.6070604@fcdonders.kun.nl> Dear Connectionists, I would to draw your attention to a paper recently published in Neuroimage: Jensen O, Hari R, Kaila K. (2002) Visually evoked gamma responses in the human brain are enhanced during voluntary hyperventilation. Neuroimage 15:575-586. The paper is a combined MEG and theoretical study on visually evoked gamma responses. By analysing a network of Hodgkin-Huxley type neurons we aim to account for properties of the visually evoked gamma response. Please contact me for reprints or visit my website: http://baserv.uci.kun.nl/~olejen Best regards, Ole Jensen ==================================================================== Ole Jensen, Ph.D. F.C. Donders Centre for Cognitive Neuroimaging P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Office : +31 24 36 10884 Mobile : +31 62 18 57495 Fax : +31 24 36 10652 e-mail : ole.jensen at fcdonders.kun.nl URL : http://baserv.uci.kun.nl/~olejen http://www.fcdonders.kun.nl From marcus at idsia.ch Thu Apr 18 08:14:48 2002 From: marcus at idsia.ch (Marcus Hutter) Date: Thu, 18 Apr 2002 14:14:48 +0200 Subject: New Paper on Mutual Information Message-ID: <00ad01c1e6d2$a3094960$0fbfb0c3@ho> Dear Colleagues, I would like to announce a new paper on Mutual Information beyond the point/frequency estimate, which today appeared in the online NIPS-Proceedings http://www-2.cs.cmu.edu/~nips/2001papers/ since it is possibly of general interest. Best wishes Marcus Hutter =========================================================================== "Distribution of Mutual Information" The mutual information of two random variables i and j with joint probabilities t_ij is commonly used in learning Bayesian nets as well as in many other fields. The chances t_ij are usually estimated by the empirical sampling frequency n_ij/n leading to a point estimate I(n_ij/n) for the mutual information. To answer questions like ``is I(n_ij/n) consistent with zero?'' or ``what is the probability that the true mutual information is much larger than the point estimate?'' one has to go beyond the point estimate. In the Bayesian framework one can answer these questions by utilizing a (second order) prior distribution p(t) comprising prior information about t. From the prior p(t) one can compute the posterior p(t|n), from which the distribution p(I|n) of the mutual information can be calculated. We derive reliable and quickly computable approximations for p(I|n). We concentrate on the mean, variance, skewness, and kurtosis, and non-informative priors. For the mean we also give an exact expression. Numerical issues and the range of validity are discussed. Downloadable also at: http://www.idsia.ch/~marcus/ai/xentropy.htm in various formats. =========================================================================== Dr. Marcus Hutter, IDSIA Istituto Dalle Molle di Studi sull'Intelligenza Artificiale Galleria 2 CH-6928 Manno(Lugano) - Switzerland Phone: +41-91-6108668 Fax: +41-91-6108661 E-mail marcus at idsia.ch http://www.idsia.ch/~marcus From bengioy at IRO.UMontreal.CA Thu Apr 18 09:06:11 2002 From: bengioy at IRO.UMontreal.CA (Yoshua Bengio) Date: Thu, 18 Apr 2002 09:06:11 -0400 Subject: PhD + post-doc positions in neural network motor models Message-ID: <20020418090611.E24965@einstein.iro.umontreal.ca> ***************************************************************** Visuomotor Control and Learning in Large-Scale Biological and Artificial Neural Networks Positions are available at the graduate and postdoctoral level to work with a team of researchers who are pursuing a series of collaborative behavioral, neurophysiological and computational studies of the neural mechanisms of visuomotor control and motor skill acquisition in the human, monkey and cat, and their application to the design of autonomous adaptive neural networks. Team members include John Kalaska, Yoshua Bengio, Trevor Drew, Paul Cisek and Stephen Scott. Physiological projects are studying the functional organization and learning mechanisms in mammalian motor systems. Modelling and theoretical tools are being used to develop and evaluate the performance of different computational architectures and learning algorithms for adaptive task decomposition in large-scale biological and artificial neural networks that are inspired by the physiological studies, and to assess their potential applications in a wide range of technological domains. Candidates have the opportunity to collaborate with several team members. Candidates must have a strong background in one or more of the following areas: learning mechanisms in artificial or biological neural networks, computer science, control theory, statistical analysis, computational neuroscience, systems neuroscience, motor control, biomechanics, psychophysics. Positions are available immediately, but start times are negotiable. Open to all nationalities. Graduate-level candidates will apply for a PhD at one of three departments: Physiology at U. Montreal, Montreal, Quebec (J. Kalaska, T. Drew, P. Cisek), Computer Science at U. Montreal (Y. Bengio), or Anatomy at Queens University, Kingston, Ontario (S. Scott). Competitive stipends. Funding available for periods of up to 3-5 years. Send CV, statement of research interests, and 2 letters of reference to John Kalaska, Dpartement de Physiologie, Pavillon Paul-G-Desmarais, Universit de Montral, 2900 Edouard-Montpetit, Montral Qubec, Canada, H3T 1J4 (telephone: 514-343-6349; fax: 514-343-6113; e-mail: kalaskaj at physio.umontreal.ca). ***************************************************************** -- Yoshua Bengio Associate professor / Professeur agrg Canada Research Chair in Statistical Learning Algorithms / titulaire de la chaire de recherche du Canada en algorithmes d'apprentissage statistique Dpartement d'Informatique et Recherche Oprationnelle Universit de Montral, adresse postale: C.P. 6128 Succ. Centre-Ville, Montral, Qubec, Canada H3C 3J7 adresse civique: 2920 Chemin de la Tour, Montral, Qubec, Canada H3T 1J8, #2194 Tel: 514-343-6804. Fax: 514-343-5834. Bureau 3339. http://www.iro.umontreal.ca/~bengioy http://www.iro.umontreal.ca/~lisa From dirk at bioss.ac.uk Fri Apr 19 04:10:56 2002 From: dirk at bioss.ac.uk (Dirk Husmeier) Date: Fri, 19 Apr 2002 09:10:56 +0100 Subject: PhD position in Bioinformatics Message-ID: <3CBFD110.B7E87015@bioss.ac.uk> A three-year PhD studentship is available at Biomathematics & Statistics Scotland (BioSS) for a highly motivated candidate to work on the application of machine learning techniques and graphical models in bioinformatics with a focus on phylogenetics and/or gene expression data analysis. The work will be carried out at the science campus of EdinburghUniversity, and is available immediately. Applicants should have, or shortly expect to obtain, a first or upper second class degree in mathematics, physics, statistics, mathematical biology or a related discipline, and strong programming skills in C/C++ and MATLAB. A familiarity with multivariate statistics, Bayesian methods, probabilistic graphical models, and numerical optimization (gradient ascent, EM algorithm, simulated annealing etc.) would be useful. Awards include full university fees (for EU nationals) and a maintenance allowance of =A310,000 per year. Please send a CV and the names and addresses of two referees to: Dirk Husmeier Biomathematics & Statistics Scotland By email: dirk at bioss.ac.uk By surface mail, until 30 April: SCRI, Inergowrie DUNDEE, DD2 5DA United Kingdom By surface mail, after 1 May: JCMB, The King's Buildings, EDINBURGH, EH9 3J United Kingdom All applications will be considered until the position has been filled. An outline of two possible projects is available at: http://www.bioss.sari.ac.uk/~dirk/temp/PhD_project.html -- ---------------------------------------------- Dirk Husmeier Biomathematics and Statistics Scotland (BioSS) SCRI, Invergowrie, Dundee DD2 5DA, United Kingdom http://www.bioss.ac.uk/~dirk/ From rens at science.uva.nl Fri Apr 19 06:48:39 2002 From: rens at science.uva.nl (Rens Bod) Date: Fri, 19 Apr 2002 12:48:39 +0200 (MET DST) Subject: Memory-based Models of Music Message-ID: Hi, The following paper-in-press may be of interest to the readers of this list: R. Bod. "Memory-Based Models of Melodic Analysis: Challenging the Gestalt Principles". Journal of New Music Research 30(3). See http://turing.wins.uva.nl/~rens/jnmr01.pdf Best wishes, Rens Bod (http://turing.wins.uva.nl/~rens) From dayan at gatsby.ucl.ac.uk Mon Apr 22 06:23:01 2002 From: dayan at gatsby.ucl.ac.uk (Peter Dayan) Date: Mon, 22 Apr 2002 11:23:01 +0100 Subject: PhD studentship with Wolfram Schultz and Peter Dayan Message-ID: <20020422112301.A3358@flies.gatsby.ucl.ac.uk> PhD Studentship Neuroscience Applications are invited for a research studentship on empirical and theoretical study of the neural basis of reinforcement learning, tenable as part of a collaboration between Wolfram Schultz, at the Department of Anatomy, University of Cambridge, and Peter Dayan, at the Gatsby Computational Neuroscience Unit, UCL. The Cambridge lab has long-standing experience in the recording and analysis of the activities of dopamine cells and their targets; the Gatsby Unit is one of the largest labs in the world devoted to computational neuroscience. Applicants should have a strong analytical background, a keen interest in neuroscience, and a good honours degree in a relevant subject. The post will likely be based in Cambridge, though substantial interaction with the Gatsby Unit in London (90 minutes away) will be required. Applications should include a CV, a statement of research interests, and a list of three referees, and should be made by 31st May 2002 by email to admissions at gatsby.ucl.ac.uk Further information about Wolfram Schultz's lab can be found at http://www.anat.cam.ac.uk/pages/staff/academic/schultz/index.html and about the Gatsby Unit at http://www.gatsby.ucl.ac.uk From jose at psychology.rutgers.edu Mon Apr 22 07:04:39 2002 From: jose at psychology.rutgers.edu (Stephen Hanson) Date: Mon, 22 Apr 2002 07:04:39 -0400 Subject: Postdoctural Position--Please REPLY in subject line RUMBA POSTDOC Message-ID: <3CC3EE47.10202@psychology.rutgers.edu> COGNITIVE/COMPUTATIONAL NEUROSCIENCE POSTDOCTORAL POSITION at RUTGERS UNIVERSITY, Newark Campus. The Rutgers University Mind/Brain Analysis (RUMBA) Project anticipates making one postdoctoral appointment, which is to begin in the SUMMER (June/July) of 2002. This positions are for a minimum of 2 years, with the possibility of continuation for 1 more year and will be in the areas of specialization of cognitive neuroscience with emphasis on the development of new paradigms and methods in neuroimaging, mathematical modeling, signal processing or data analysis in functional brain imaging. Particular interest is in methods and algorithms for fusion of EEG/fMRI. Applications are welcomed begining immediately and review will continue until the position is filled. Rutgers University is an equal opportunity/affirmative action employer. Qualified women and minority candidates are especially encouraged to apply. Send CV and three letters of recommendation and 1 reprint to Professor S.J. Hanson, Department of Psychology, Rutgers University, Newark, NJ 07102. Email enquiry can be made to jose at psychology.rutgers.edu please put "RUMBA POSTDOC" in your subject field also see http://www.rumba.rutgers.edu. From frank.pasemann at ais.fraunhofer.de Tue Apr 23 12:05:49 2002 From: frank.pasemann at ais.fraunhofer.de (Frank Paseman) Date: Tue, 23 Apr 2002 18:05:49 +0200 Subject: Research Associate - Evolutionary Robotics Message-ID: <3CC5865D.20980EAE@ais.fraunhofer.de> I would be grateful if you could distribute the following announcement: The Fraunhofer Institute for Autonomous Intelligent Systems, Sankt Augustin, Germany, invites applications for a Research Associate position in the area of evolutionary robotics. The project will focus on the evolution of modular neurocontrollers for autonomous systems combined with the implementation of behavior relevant learning rules for general recurrent neural networks. Experiments will involve different robot platforms like Kheperas, quadruped and hexapod robots, holonomic robots, etc. The more theoretically oriented activities will be concerned with the relationship between robot behavior, network structure, and network dynamics, theory of evolutionary processes and optimization of evolutionary search strategies, behavior oriented on-line learning rules. The ideal candidate should have strong background in one or more of the following areas: dynamical systems theory, complex systems theory, control theory, recurrent neural networks, behavior based robotics, embodied cognitive science. He should have strong programming skills in C/C++ under a Linux environment. Experience with programming simulators for robots and physical environments will be an advantage. Applicants should have a university degree (Masters or german diploma) in computer science, physics or mathematics. Funding is available for 3 years. The salary will be on the BAT-IIa scale. The post is tenable immediately, but start time is negotiable. For further information and informal enquiries please contact Prof. F. Pasemann (frank.pasemann at ais.fraunhofer.de/http://www.ais.fraunhofer.de/INDY). ************************************************************** Prof. Dr. Frank Pasemann Tel: x49-2241-142373 Fraunhofer-Institut Fax: x49-2241-142342 Autonome Intelligente Systeme Schloss Birlinghoven frank.pasemann at ais.fraunhofer.de D-53754 Sankt Augustin http://www.ais.fraunhofer.de/INDY From becker at meitner.psychology.mcmaster.ca Wed Apr 24 09:28:31 2002 From: becker at meitner.psychology.mcmaster.ca (S. Becker) Date: Wed, 24 Apr 2002 09:28:31 -0400 (EDT) Subject: predoc/doctoral/postdoc positions in auditory neuroscience Message-ID: RESEARCH POSITIONS AND TRAINING OPPORTIUNITES IN AUDITORY MODELLING, PSYCHOPHYSICAL TESTING, AND PLASTICITY IN THE AUDITORY SYSTEM AT MCMASTER UNIVERSITY We are seeking candidates for the following three positions in human auditory neuroscience at McMaster University. Students interested in PhD training in the these fields are also invited to contact us. (1) COMPUTATIONAL MODELS OF HEARING LOSS Postdoctoral and doctoral candidates are sought to develop computational models of sub-cortical and cortical reorganization after noise-induced hearing loss. Both positions will be based at McMaster, in the departments of Psychology and Electrical and Computer Engineering. Excellent programming skills are required. Experience with neural network modelling and signal processing and knowledge of the auditory system and psychoacoustics are highly desirable. The postdoctoral position requires a PhD in Cognitive Science, Computer Science, Electrical Engineering or a related discipline. This research is part of a multi-disciplinary project to study mechanisms and treatment of tinnitus, funded by a $1.1M CIHR grant (NET Programme) to L. Roberts and S. Becker (Psychology, McMaster University), I. Bruce (Electrical and Computer Engnineering, McMaster University), J. Eggermont (Faculty of Medicine, University of Calgary), C. Pantev (Rotman Research Institute, Toronto) and L. Ward (Psychology, University of British Columbia). Interested candidates should send a letter of intention and CV via email with subject line "tinnitus postdoc/doctoral positions" to both Dr. S. Becker (becker at mcmaster.ca) and Dr. I. Bruce (ibruce at ieee.org). (2) PSYCHOPHYSICAL ASSESSMENT OF AUDITORY FUNCTION A Research Associate is sought to conduct psychophysical tests of speech intelligibility in normal and hearing-impaired participants, in order to evaluate hearing compensation algorithms. Experience in psychophysical testing, computer programming, and signal processing are highly desirable. This work is part of a collaborative project to develop intelligent signal processing and neural network algorithms for hearing aids, funded by a $0.5M Collaborative Research Opportunities grant from NSERC, Canada to a multi-disciplinary team of researchers at McMaster University: S. Haykin (Electrical and Computer Engineering), L. Trainor, S. Becker. R. Racine and J. Platt (Psychology). Interested candidates should send a letter of intention and CV via email with subject line "psychoacoustics RA position" to Dr. L. Trainor, ljt at mcmail.cis.mcmaster.ca (3) NEURAL PLASTICITY IN THE AUDITORY SYSTEM We are seeking a scientific colleague to work with us on evaluating models of residual inhibition in tinnitus and on EEG studies of neural plasticity in the human auditory system. The candidate should be a self-directed individual who can work effectively in a collaborative research environment. Strong computing skills and a working knowledge of the auditory system and signal processing are important assets. This position is available at the postdoctoral or similar level, although senior PhD students with suitable background may be considered. The research is part of a multi-site CIHR NET Progamme on understanding, treating, and preventing tinnitus, which is based at McMaster University and three other institutions (see position #1 above). Interested candidates should send their curriculum vitas and statement of interest by email to Dr. L.E. Roberts (roberts at mcmaster.ca). PLease put the phrase "Auditory Scientist Position" in the subject line of your email message. For further information on the research interests of the team see: Tinnitus project: www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.psychology.mcmaster.ca/hnplab www.ucalgary.ca/UofC/faculties/medicine/PHBI/faculty/eggermont.html www.rotman-baycrest.on.ca/content/people/profiles/pantev.html http://neuron2.psych.ubc.ca/~lward/people/index.html Intelligent hearing aids project: www.science.mcmaster.ca/Psychology/http://www.psychology.mcmaster.ca/ljt/ www.crl.mcmaster.ca/People/Faculty/Haykin/haykin.html www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.science.mcmaster.ca/Psychology/racine-web-page/racine.html www.science.mcmaster.ca/Psychology/platt/jp.html From bogus@does.not.exist.com Wed Apr 24 04:20:44 2002 From: bogus@does.not.exist.com () Date: Wed, 24 Apr 2002 01:20:44 -0700 Subject: AI Statistics Conference, Key West, January 2003: Preliminary call for papers Message-ID: <3C0C2AFA8DA9F5448E401726A56637F6AE56E7@RED-MSG-10.redmond.corp.microsoft.com> PRELIMINARY CALL FOR PAPERS Ninth International Conference on Artificial Intelligence and Statistics January 3-6, 2003, Hyatt Hotel, Key West, Florida http://research.microsoft.com/conferences/AIStats2003/ This is the ninth in a series of workshops which have brought together researchers in Artificial Intelligence and in Statistics to discuss problems of mutual interest. The exchange has broadened research in both fields and has strongly encouraged interdisciplinary work. Papers on all aspects of the interface between AI & Statistics are encouraged. Deadlines: Submission of full length 4-page papers for review: 2 August 2002 Submission of final camera-ready papers: 1 November 2002 Invited Speakers: Andrew Blake (Microsoft Research, Cambridge) Bill Freeman (MIT) Zoubin Ghahramani (Gatsby Computational Neuroscience Unit) David Haussler (UCSC) Geoffrey Hinton (University of Toronto) Tommi Jaakkola (MIT) Larry Saul (University of Pennsylvania) Key West provides a superb location for this workshop, and the weather at Key West in January is expected to be very pleasant. The workshop timetable will focus on morning and early evening sessions, allowing ample free time in the afternoons for scientific discussions or to take advantage of local attractions such as scuba diving, snorkelling, wave runners, parasailing, fishing, sailing and golf. From rich at richsutton.com Thu Apr 25 18:53:22 2002 From: rich at richsutton.com (Rich Sutton) Date: Thu, 25 Apr 2002 18:53:22 -0400 Subject: RL [Reinforcement Learning] FAQ Message-ID: Dear Connectionists, I have made a basic RL FAQ available at http://richsutton.com/RL-FAQ.html. Additions/corrections/improvements are encouraged. Rich Sutton From cindy at bu.edu Thu Apr 25 15:34:38 2002 From: cindy at bu.edu (Cynthia Bradford) Date: Thu, 25 Apr 2002 15:34:38 -0400 Subject: 6th ICCNS: Call for Registration Message-ID: <200204251934.g3PJYcD04549@cns-pc75.bu.edu> Apologies if you receive this more than once. ***** CALL FOR REGISTRATION ***** ***** AND ***** ***** FINAL INVITED PROGRAM ***** SIXTH INTERNATIONAL CONFERENCE ON COGNITIVE AND NEURAL SYSTEMS Tutorials: May 29, 2002 Meeting: May 30 - June 1, 2002 Boston University http://www.cns.bu.edu/meetings/ This interdisciplinary conference focuses on two fundamental questions: How Does the Brain Control Behavior? How Can Technology Emulate Biological Intelligence? A single oral or poster session enables all presented work to be highly visible. Contributed talks will be presented on each of the three conference days. Three-hour poster sessions with no conflicting events will be held on two of the conference days. All posters will be up all day, and can also be viewed during breaks in the talk schedule. CONFIRMED INVITED SPEAKERS TUTORIAL SPEAKERS: Wednesday, May 29, 2002 Mark Gluck (Rutgers University) Neural networks in neurology and clinical neuropsychology: Alzheimer's disease, amnesia, and Parkinson's disease Gail A. Carpenter (Boston University) Adaptive resonance theory Ferdinando Mussa-Ivaldi (Northwestern University Medical School) Learning and adaptive control of arm movements Frank Guenther (Boston University) A model of the neural bases of speech motor control INVITED SPEAKERS Thursday, May 30, 2002 CELL AND CIRCUIT DYNAMICS: Daniel Johnston (Baylor College of Medicine) Information processing and storage by neuronal dendrites Bard Ermentrout (University of Pittsburgh) The role of oscillations in odor learning in the Limax John Rinzel (New York University) Cellular dynamics involved in sound localization VISION AND IMAGE PROCESSING: Rudiger von der Heydt (Johns Hopkins University School of Medicine) Global structure in local feature maps David J. Field (Cornell University) Visual systems and the statistics of natural scenes: How far can we go? Philip J. Kellman (UCLA) From jebara at cs.columbia.edu Fri Apr 26 21:55:55 2002 From: jebara at cs.columbia.edu (Tony Jebara) Date: Fri, 26 Apr 2002 18:55:55 -0700 Subject: ICDL'02 - Development and Learning Message-ID: Dear Connectionists, You may be interested in attending the following conference. Please note that early registration rates are available until May 5th, 2002. Details and the program are available at: http://www.egr.msu.edu/icdl02/ Apologies if you receive multiple copies of this message. Thanks... -Tony Jebara ------------------------------------------------------------------------- The 2nd International Conference on Development and Learning (ICDL'02) June 12 - 15, 2002 Massachusetts Institute of Technology Cambridge, Massachusetts USA http://www.egr.msu.edu/icdl02/ Sponsored by: American Association for Artificial Intelligence Cognitive Science Society IEEE Computer Society IEEE Neural Networks Society IEEE Robotics and Automation Society Recent advances in artificial intelligence, cognitive science, neuroscience and robotics have stimulated the birth and growth of a new research field, known as computational autonomous mental development. Although human mental development is a well known subject of study, e.g., in developmental psychology, computational studies of mental development for either machines or humans had not received sufficient attention in the past. Mental development is a process during which a brain-like natural or artificial embodied system, under the control of its intrinsic species-specific developmental program residing in the genes or artificially designed, develops mental capabilities through its autonomous real-time interactions with its environments (including its own internal environment and components) using its own sensors and effectors. The scope of mental development includes cognitive, behavioral, emotional and all other mental capabilities that are exhibited by humans, higher animals and artificial systems. Investigations of the computational mechanisms of mental development are expected to improve our systematic understanding of the working of the wide variety of cognitive and behavioral capabilities in humans and to enable autonomous development of these highly complex capabilities by robots and other artificial systems. Organizers: James L. McClelland Alex P. Pentland Jeff Elman Mriganka Sur Juyang Weng Sridhar Mahadevan Tony Jebara Program Overview: June 12, 2002: Cortical Development and Learning during Vision, Recognition, and Action (Stephen Grossberg) Autonomous Mental Development for Robots (Juyang Weng) June 13, 2002: Cortical Plasticity Contributing to Variations in Human Performance Ability (Michael Merzenich) A Developing Sensory Mapping for Robots (N. Zhang, J. Weng, and Z. Zhang) Combining Embodied Models and Empirical Research for Understanding the Development of Shared Attention (I. Fasel, G. Deak, J. Triesch, and J. Movellan) Learning to Recognize Human Action Sequences (C. Yu, and D. Ballard) The Development of Gaze Following as a Bayesian Systems Identification Problem (J. Movellan, and J.S. Watson) Prediction-error Driven Learning: The Engine of Change in Cognitive Development (James McClelland) Combining Configural and TD Learning on a Robot (D.S. Touretzky, N.D. Daw, and E.J. Tira-Thompson) Action Chaining by a Developmental Robot with a Value System (Y. Zhang, and J. Weng) A Model of an Expectancy-driven and Analogy-making Actor (A. Kulakov, G. Stojanov, and D. Davcec) How do Features of Sensory Representations Develop? (Jon Kass) Dav: A Humanoid Robot Platform for Autonomous Mental Development (J.D. Han, S.W. Zeng, K.Y. Tham, M. Badgero, and J. Weng) A Novel Optimal Discriminant Principle in High Dimensional Spaces (Y. Guo, and L. Wu) Cortical Software Re-Use: A Computational Principle for Cognitive Robotics (R. Reilly, and I. Marian) Contentful Mental States for Robot Baby (P.R. Cohen, J.T. Oates, and C.R. Beal) An Incremental Representation of Conceptual Symbols Using RCE Neural Network (M. Xie, and M.L.Yuan) A Developmental Principle for Robotic Hand-Eye Coordination Skill (M. Xie) An Evolutional Network Architecture for Developmental Knowledge Bases (A. Saad, and A.R.M. Zaghloul) Towards a Theory Grounded Theory of Language (C.G. Prince, E.J. Mislivec, O.V. Kosolapaov, and T.R. Lykken) June 14,2002: One Thing Follows Another: Initial State, Task, and Developmental Change in Human Infants (Esther Thelen) A Theory for Mentally Developing Robots (J. Weng) Many-Layered Learning (P.E. Utgoff, and D.J. Stracuzzi) Beyond the Turing Test: Performance Metrics for Evaluating a Computer Simulation of the Human Mind (N. Alvarado, S. S. Adams, S. Burbeck, and C. Latta ) Learning in Content-based Image Retrieval (Thomas S. Huang) Developmental Learning of Memory-based Perceptual Models (Y. Ivanov, and B. Blumberg) Learning to Detect Multi-View Faces in Real-Time (S.Z. Li, L. Zhu, Z.Q,. Zhang, and H.T. Zhang) Local Non-Negative Matrix Factorization as a Visual Representation (T. Feng, S.Z. Li, H.Y. Shum, and H.J. Zhang) Humanoid Robot Models of Child Development (Rodney Brooks) Statistical Imitative Learning from Perceptual Data (T. Jebara, and A. Pentland) Learning Prospective Pick and Place Behavior (D. Wheeler, A.H. Fagg, and R.A. Grupen) Learning Movement Sequences from Demonstration (A. Ramesh, and M.J. Mataric) June 15, 2002: Rewiring Cortex: Rules of Cortical Network Development (Mriganka Sur) Interactions Between Development and Learning During the Acquisition of Binocular Disparity Sensitivities (M. Dominguez, and R.A. Jacobs) Dynamic Growth Modeling of Human Cognitive Microdevelopment (Z. Yan, and K. Fischer) Marginal Self-organization: A Model of the Role of Executive Processes in Learning (R. Viviani) Learning Your Life: Wearables and Familiars (Alex Pentland) On the Development of Visual Object Memory: The Stay/Go Decision Problem (C.T. Morrison, P. Cohen, and P. Sebastiani) A Computer-based Tutoring System for Visual-Spatial Skills: Dynamically Adapting to the User?s Developmental Range (M.W. Connell, and D.A. Stevens) Extending the BDI Model to Accelerate the Mental Development of Autistic Patients (B. Galitsky) Development as a Source of Complexity (Jeff Elman) Dopamine, Reward Conditioning and Robot Behavior (O. Sporns, and W.H. Alexander) Dopamine and Inference About Timing (N. Daw, A.C. Courville, and D.S. Touretzky) A Developmental Approach Accelerates Learning of Joint Attention (Y. Nagai, M. Asada, and K. Hosoda) From laubach at kafka.med.yale.edu Sun Apr 28 21:15:02 2002 From: laubach at kafka.med.yale.edu (Mark Laubach) Date: Sun, 28 Apr 2002 21:15:02 -0400 Subject: PROGRAMMING THE "BRAIN-MACHINE INTERFACE" Message-ID: PROGRAMMING THE "BRAIN-MACHINE INTERFACE" A position is available at the John B. Pierce Laboratory, affiliated with the Yale School of Medicine, for someone with strong computer programming skills (C/C++, Python, Matlab, S+/R). This person will work as part of a team of neuroscientists, engineers, mathematicians, and computer scientists working on the "brain-machine interface" (sponsored by a grant from DARPA). This position requires knowledge of UNIX-style operating systems, modern methods for data analysis and machine learning, and distributed and/or real-time computing, is available starting July 1, 2002, and is paid at NIH-standard levels. If interested, please send a vita, a summary of skills and experience, and the names of three references to: Mark Laubach, Ph.D. The John B. Pierce Laboratory Yale School of Medicine 290 Congress Ave New Haven CT 06519 E-mail: laubach at kafka.med.yale.edu http://www.jbpierce.org/staff/laubach.html ------ NOTE: This is a follow-up to a recent posting related to this project. Three of the other positions advertised in our original posting have now been filled. From mpessk at guppy.mpe.nus.edu.sg Mon Apr 29 01:00:42 2002 From: mpessk at guppy.mpe.nus.edu.sg (S. Sathiya Keerthi) Date: Mon, 29 Apr 2002 13:00:42 +0800 (SGT) Subject: Extensions of the SMO algorithm Message-ID: Dear Connectionists: We have recently completed two papers on the extensions of the SMO algorithm to Least Squares SVM formulations and Kernel Logistic Regression. Gzipped postscript files containing these papers can be downloaded from: http://guppy.mpe.nus.edu.sg/~mpessk/svm.shtml The titles and abstracts of these papers are given below. S. Keerthi -------------------------------------------------------------- SMO Algorithm for Least Squares SVM Formulations S.S. Keerthi and S.K. Shevade This paper extends the well-known SMO algorithm of Support Vector Machines (SVMs) to Least Squares SVM formulations which include LS-SVM classification, Kernel Ridge Regression and a particular form of regularized Kernel Fisher Discriminant. The algorithm is shown to be asymptotically convergent. It is also extremely easy to implement. Computational experiments show that the algorithm is fast and scales efficiently (quadratically) as a function of the number of examples. -------------------------------------------------------------- A Fast Dual Algorithm for Kernel Logistic Regression S.S. Keerthi, K. Duan, S.K. Shevade and A.N. Poo (Accepted for presentation at ICML 2002) This paper gives a new iterative algorithm for kernel logistic regression. It is based on the solution of the dual problem using ideas similar to those of the SMO algorithm for Support Vector Machines. Asymptotic convergence of the algorithm is proved. Preliminary computational experiments show that the algorithm is robust and fast. The algorithmic ideas can also be used to give a fast dual algorithm for solving the optimization problem arising in the inner loop of Gaussian Process classifiers. -------------------------------------------------------------- From amy at cs.umass.edu Mon Apr 29 17:39:10 2002 From: amy at cs.umass.edu (Amy McGovern) Date: Mon, 29 Apr 2002 17:39:10 -0400 Subject: Ph.D. thesis available Message-ID: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Dear Connectionists, I am pleased to announce the availability of my Ph.D. thesis: Autonomous Discovery of Temporal Abstractions from Interaction with an Environment Amy McGovern University of Massachusetts Amherst The thesis is available from either of the following links: http://www-anw.cs.umass.edu/~amy/pubs/mcgovern_thesis.pdf http://www-anw.cs.umass.edu/~amy/pubs/mcgovern_thesis.ps --------------------------------- Abstract The ability to create and to use abstractions in complex environments, that is, to systematically ignore irrelevant details, is a key reason that humans are effective problem solvers. Although the utility of abstraction is commonly accepted, there has been relatively little research on autonomously discovering or creating useful abstractions. A system that can create new abstractions autonomously can learn and plan in situations that its original designer was not able to anticipate. This dissertation introduces two related methods that allow an agent to autonomously discover and create temporal abstractions from its accumulated experience with its environment. A temporal abstraction is an encapsulation of a complex set of actions into a single higher-level action that allows an agent to learn and plan while ignoring details that appear at finer levels of temporal resolution. The main idea of both methods is to search for patterns that occur frequently within an agent's accumulated successful experience and that do not occur in unsuccessful experiences. These patterns are used to create the new temporal abstractions. The two types of temporal abstractions that our methods create are 1) subgoals and closed-loop policies for achieving them, and 2) open-loop policies, or action sequences, that are useful ``macros.'' We demonstrate the utility of both types of temporal abstractions in several simulated tasks, including two simulated mobile robot tasks. We use these tasks to demonstrate that the autonomously created temporal abstractions can both facilitate the learning of an agent within a task and can enable effective knowledge transfer to related tasks. As a larger task, we focus on the difficult problem of scheduling the assembly instructions for computers with multiple pipelines in such a manner that the reordered instructions will execute as quickly as possible. We demonstrate that the autonomously discovered action sequences can significantly improve performance of the scheduler and can enable effective knowledge transfer across similar processors. Both methods can extract the temporal abstractions from collections of behavioral trajectories generated by different processes. In particular, we demonstrate that the methods can be effective when applied to collections generated by reinforcement learning agents, heuristic searchers, and human tele-operators. From juergen at idsia.ch Tue Apr 30 07:34:58 2002 From: juergen at idsia.ch (Juergen Schmidhuber) Date: Tue, 30 Apr 2002 13:34:58 +0200 Subject: Ph.D. thesis on temporal abstraction References: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Message-ID: <3CCE8162.11399073@idsia.ch> Re: Amy McGovern's announcement of "Autonomous Discovery of Temporal Abstractions from Interaction with an Environment" Let us not forget Mark Ring's early work on temporal abstraction for reinforcement learning, e.g: Mark. B. Ring: Incremental Development of Complex Behaviors through Automatic Construction of Sensory-Motor Hierarchies. Machine Learning: Proceedings of the Eighth International Workshop, In L. Birnbaum and G. Collins, eds., 343-347, Morgan Kaufmann, 1991. Additional algorithms in his NIPS 5 paper (1993) and his 1994 thesis. Maybe also check out http://www.idsia.ch/~juergen/subgoals.html Juergen Schmidhuber http://www.idsia.ch/~juergen From h.wang at ulster.ac.uk Tue Apr 30 07:55:01 2002 From: h.wang at ulster.ac.uk (Hui Wang) Date: Tue, 30 Apr 2002 12:55:01 +0100 Subject: Special Issue of Decision Support Systems on Data Mining for Financial Decision Making Message-ID: <072701c1f03d$dbe8e870$a2943dc1@tiger> Dear Connectionists, You may be interested in the following special issue of Decision Support Systems. Details are available at http://www.weigend.com/dss and http://www.elsevier.com/inca/homepage/sae/orms/dss/call1.htm Apologies if you receive multiple copies of this message. Thanks... Hui Wang Andreas Weigend ------------------------------------------------------------------------------- The Journal of Decision Support Systems Special Issue on Data Mining for Financial Decision Making GUEST EDITORS Hui Wang, University of Ulster Andreas S. Weigend, Weigend Associates LLC CALL FOR PAPERS As information intensive organizations transform themselves from passive collectors to active explorers and exploiters of data, they face a serious challenge: How can they benefit from increased access to information to better understand their markets, customers, suppliers, operations and internal business processes? Responding to this challenge, the field of data mining has emerged. It focuses on the process of discovering valid, comprehensible, and potentially useful knowledge from large data sets with the goal to apply this knowledge to decision making. Data mining integrates concepts from modern statistics, intelligent information systems, machine learning, pattern recognition, decision theory, data engineering and database management, and provides powerful tools that can reveal complex and hidden relationships in large amounts of data. The approaches include neural networks, genetic programming, and tree-based methods. Data mining already has a major impact on business and finance. Financial markets generate large volumes of data. Analysing these data to reveal valuable information and making use of the information in decision making present great opportunities but grand challenges for data mining. The rewards for finding valuable patterns are potentially enormous, but so are the difficulties. There is evidence that short-term trends do exist and some general patterns do occur frequently. Important problems are: how to find the trends at their early stages and how to time the beginning and ending of trends, how to take into account in decision making the found trends, the general patterns, and domain knowledge that describes the intricately inter-related world of global financial markets. The focus of this special issue is on the use of data mining techniques for decision making in financial markets. Topics of interest include: * Financial data selection and pre-processing for data mining * Solutions to new problems in financial decision making * New solutions for classical problems in financial decision making * Data and solutions visualisation for financial decision making * Successful case studies. Areas include: * Risk management including credit risk and market risk * Asset allocation, dynamic trading and hedging * Execution and liquidity models * Behavioural finance, and other emerging areas. Both original contributions and thoughtful survey papers are welcome. SUBMISSION INSTRUCTIONS Electronic submissions are strongly encouraged. Postscript or PDF copies of manuscripts may be emailed to h.wang at ulst.ac.uk. SUBMISSION DEADLINE: September 9, 2002 Details about the submission process and scope of the special issue are available at http://www.weigend.com/dss and http://www.elsevier.com/inca/homepage/sae/orms/dss/call1.htm Hui Wang School of Information and Software Engineering University of Ulster at Jordanstown Northern Ireland, BT37 0QB United Kingdom Tel: +44 28 90368981 Fax: +44 28 90366068 Email: h.wang at ulst.ac.uk Andreas S. Weigend Weigend Associates LLC P.O.Box 20207 Stanford, CA 94309 U.S.A. Tel: +1 917 697-3800 Fax: +1 815 327-5462 Email: dss at weigend.com From juergen at idsia.ch Tue Apr 30 07:34:58 2002 From: juergen at idsia.ch (Juergen Schmidhuber) Date: Tue, 30 Apr 2002 13:34:58 +0200 Subject: Ph.D. thesis on temporal abstraction References: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Message-ID: <3CCE8162.11399073@idsia.ch> Re: Amy McGovern's announcement of "Autonomous Discovery of Temporal Abstractions from Interaction with an Environment" Let us not forget Mark Ring's early work on temporal abstraction for reinforcement learning, e.g: Mark. B. Ring: Incremental Development of Complex Behaviors through Automatic Construction of Sensory-Motor Hierarchies. Machine Learning: Proceedings of the Eighth International Workshop, In L. Birnbaum and G. Collins, eds., 343-347, Morgan Kaufmann, 1991. Additional algorithms in his NIPS 5 paper (1993) and his 1994 thesis. Maybe also check out http://www.idsia.ch/~juergen/subgoals.html Juergen Schmidhuber http://www.idsia.ch/~juergen From mpessk at guppy.mpe.nus.edu.sg Mon Apr 29 01:00:42 2002 From: mpessk at guppy.mpe.nus.edu.sg (S. Sathiya Keerthi) Date: Mon, 29 Apr 2002 13:00:42 +0800 (SGT) Subject: Extensions of the SMO algorithm Message-ID: Dear Connectionists: We have recently completed two papers on the extensions of the SMO algorithm to Least Squares SVM formulations and Kernel Logistic Regression. Gzipped postscript files containing these papers can be downloaded from: http://guppy.mpe.nus.edu.sg/~mpessk/svm.shtml The titles and abstracts of these papers are given below. S. Keerthi -------------------------------------------------------------- SMO Algorithm for Least Squares SVM Formulations S.S. Keerthi and S.K. Shevade This paper extends the well-known SMO algorithm of Support Vector Machines (SVMs) to Least Squares SVM formulations which include LS-SVM classification, Kernel Ridge Regression and a particular form of regularized Kernel Fisher Discriminant. The algorithm is shown to be asymptotically convergent. It is also extremely easy to implement. Computational experiments show that the algorithm is fast and scales efficiently (quadratically) as a function of the number of examples. -------------------------------------------------------------- A Fast Dual Algorithm for Kernel Logistic Regression S.S. Keerthi, K. Duan, S.K. Shevade and A.N. Poo (Accepted for presentation at ICML 2002) This paper gives a new iterative algorithm for kernel logistic regression. It is based on the solution of the dual problem using ideas similar to those of the SMO algorithm for Support Vector Machines. Asymptotic convergence of the algorithm is proved. Preliminary computational experiments show that the algorithm is robust and fast. The algorithmic ideas can also be used to give a fast dual algorithm for solving the optimization problem arising in the inner loop of Gaussian Process classifiers. -------------------------------------------------------------- From giro-ci0 at wpmail.paisley.ac.uk Mon Apr 1 17:09:42 2002 From: giro-ci0 at wpmail.paisley.ac.uk (Mark Girolami) Date: Mon, 01 Apr 2002 23:09:42 +0100 Subject: Research Positions Message-ID: Post-Doctoral Research Assistantship A research project which is funded by the Engineering and Physical Sciences Research Council (EPSRC), the Department of Trade and Industry (DTI), and industrial partners will be conducted at the University of Paisley in Scotland for a period of three years. The project aims to develop the technology required in software systems which will be able to provide effective detection and subsequent analysis of fraudulent activity within the general framework required of emerging fixed and mobile telecommunications applications such as electronic and mobile commerce. One postdoctoral position is now available to investigate the application of machine learning and advanced data mining methods in the detection and analysis of anomalous and possibly fraudulent usage of fixed and mobile telecommunications applications such as electronic and mobile commerce. The project will involve the design and implementation of novel algorithms and systems to discover and analyse emerging patterns of anomalous telecommunication system user activity. Highly motivated candidates who have a publication record in, ideally, machine learning, data mining or artificial intelligence applications are encouraged to apply. Applicants should have, or shortly expect to obtain, a PhD in Computer Science. Salaries will be on the R1A scale, starting at 22,200pa to 27,550pa. For further information and informal enquiries please contact Mark Girolami (mark.girolami at paisley.ac.uk). EPSRC & DTI Project Data Mining Tools for Fraud Detection in M-Commerce * DETECTOR http://cis.paisley.ac.uk/giro-ci0/projects.html Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Legal disclaimer -------------------------- The information transmitted is the property of the University of Paisley and is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination and other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer. -------------------------- From terry at salk.edu Tue Apr 2 13:33:59 2002 From: terry at salk.edu (Terry Sejnowski) Date: Tue, 2 Apr 2002 10:33:59 -0800 (PST) Subject: NEURAL COMPUTATION 14:5 Message-ID: <200204021833.g32IXx834531@purkinje.salk.edu> Neural Computation - Contents - Volume 14, Number 5 - May 1, 2002 ARTICLE A Population Study of Integrate-and-Fire-or-Burst Neurons A.R.R. Casti, A. Omurtag, A. Sornborger, E. Kaplan, B. Knight, J. Victor and L. Sirovich NOTE Stable Propagation of Activity Pulses in Populations of Spiking Neurons Werner M. Kistler and Wulfram Gerstner LETTERS Population Coding and Decoding in a Neural Field: A Computational Study Si Wu, Shun-ichi Amari, and Hiroyuki Nakahara The Influence of the Limit Cycle Topology on the Phase Resetting Curve Sorinel A. Oprisan and Carmen C. Canavier Fields As Limit Functions of Stochastic Discrimination and Their Adaptability Philip Van Loocke Learning to Recognize Three-Dimensional Objects Dan Roth, Ming-Hsuan Yang, and Narendra Ahuja A Parallel Mixture of SVMs for Very Large Scale Problems Ronan Collobert, Samy Bengio and Yoshua Bengio Bayesian Framework for Least-Squares Support Vector Machine Classifiers, Gaussian Processes and Kernel Fisher Discriminant Analysis Tony Van Gestel, Johan A.K. Suykens, Gert Lanckriet, Annemie Lambrechts, Bart De Moor and Joos Vandewalle Neural Network Pruning with Tukey-Kramer Multiple Comparison Procedure Donald E. Duckro, Dennis W. Quinn, and Samuel J. Gardner III Products of Gaussians and Probabilistic Minor Component Analysis Chris Williams and Felix Agakov Optimization of the Kernel Functions in a Probabilistic Neural Network Analyzing the Local Pattern Distribution I. Galleske and J. Castellanos Methods for Binary Multidimensional Scaling Douglas L. T. Rohde ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2002 - VOLUME 14 - 12 ISSUES USA Canada* Other Countries Student/Retired $60 $64.20 $108 Individual $88 $94.16 $136 Institution $506 $451.42 $554 * 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 rebollol at aston.ac.uk Wed Apr 3 11:13:41 2002 From: rebollol at aston.ac.uk (laura rebollo-neira) Date: Wed, 03 Apr 2002 16:13:41 +0000 Subject: post doctoral position Message-ID: <3CAB2A35.141E13AB@aston.ac.uk> I would be grateful if you could distribute the following announcement: ASTON UNIVERSITY- SCHOOL OF ENGINEERING & APPLIED SCIENCE RESEARCH ASSOCIATE/FELLOW (3 years fixed term) Starting salary 19.661 pounds Applications are invited for a Post Doctoral Research Associate to work on a 3 years EPSRC funded project entitled "Biorthogonal techniques for optimal signal representation" The project will focus on developing a framework for signal representation outside the traditional orthogonal basis setting, with applications to different areas of signal processing. Further details on the research project can be found in http://www.ncrg.aston.ac.uk/Research.html by clicking on the link for "Biorthogonal techniques for optimal signal representation" Applicants should hold, or expect to hold, a PhD in Physics, Mathematics, Computer Science, Engineering or related discipline. The work requires good mathematical and computational skills and strong interest in some area of signal processing. The successful candidate will have the opportunity to join the the well established and internationally recognized research group NCRG http://www.ncrg.aston.ac.uk The post is tenable from 1 July or as soon as possible thereafter. Informal enquiries are welcomed and should be directed to Dr Laura Rebollo-Neira: rebollol at aston.ac.uk or Prof David Lowe: d.lowe at aston.ac.uk Further particulars and application forms may be obtained electronically on http://www.aston.ac.uk/hr/recruitment.htm or from Personnel Office on Tel (0) 121 359 0870 email b.a.power at aston.ac.uk quoting Ref R02/51 Closing date for applications: 3 May 2002 ~ From gcc at sys.uea.ac.uk Thu Apr 4 07:40:56 2002 From: gcc at sys.uea.ac.uk (Gavin Cawley) Date: Thu, 04 Apr 2002 13:40:56 +0100 Subject: Postdoctoral Research Associate and PhD studentship in Bioinformatics/Machine Learning Message-ID: <5.1.0.14.0.20020404133016.020e38e0@mail.sys.uea.ac.uk> The Exploiting Genomics Programme of the BBSRC and EPSRC has funded a joint project in the John Innes Centre, University of East Anglia and Nottingham University to establish and validate methods to interpret promoter function and establish gene regulatory networks in the model plant Arabidopsis thaliana. A combination of informatics and experimental approaches will be taken using the complete genome sequence, comprising 25,000 annotated genes and extensive full-length cDNA sequence. Dr Gavin Cawley of the School of Information Systems at the University of East Anglia will use bioinformatics and machine learning (including Support Vector Machines) to analyse and classify Arabidopsis promoters using gene expression data from microarrays and gene trap insertion populations. Activities at the UEA require a full time postdoctoral research associate (to be appointed on the Research and Analogous 1A scale, starting salary 20,470 per annum). A further part-time (50%) appointment is available for a postgraduate research associate, who must also register for a PhD. Starting salary will be 9,327 per annum (pro rata to 18,655 per annum) on the Research and Analogous 1B scale. Both posts are for a fixed term of 36 months. Applicants with interests in machine learning (especially kernel methods) and/or bioinformatics are encouraged to apply. Informal enquiries may be made to Dr Cawley by e-mail gcc at sys.uea.ac.uk Applications (stating whether you are applying for the postdoctoral or postgraduate position) in the form of a letter of application, Curriculum Vitae, and the names and addresses of three referees should be sent to Dr Gavin Cawley, School of Information Systems, University of East Anglia, Norwich NR4 7TJ, UK, by no later 19 April 2002. Dr Gavin Cawley School of Information Systems University of East Anglia Norwich NR4 7TJ United Kingdom From becker at nips.salk.edu Mon Apr 8 17:13:46 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:13:46 -0700 (PDT) Subject: NIPS*2002 call for papers Message-ID: __________________________________________________________________ CALL FOR PAPERS Neural Information Processing Systems Natural and Synthetic Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada _________________________________________________________________ Submissions are solicited for the sixteenth meeting of an interdisciplinary conference, which brings together cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians 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. It is single track and highly selective. Preceding the main conference will be one day of tutorial presentations (Dec.9), and following it there will be two days of focused workshops on topical issues at Whistler/Blackcomb ski resort (Dec.13-14). NIPS*2002 INVITED SPEAKERS: Hugh Durrant-Whyte, University of Sydney: Information flow in sensor networks; Paul Glimcher, New York University: Decisions, Uncertainty and the Brain: Neuroeconomics; Deborah Gordon, Stanford University: Ants at Work; Andrew W. Moore, Carnegie Mellon University: Statistical Data Mining; Pietro Perona, Caltech: Learning visual categories; David Heeger, Stanford: Title TBA NIPS*2002 TUTORIAL SPEAKERS: Martin Cooke: Computational auditory scene analysis in listeners and machines; Richard M. Karp: Mathematical, Statistical and Algorithmic Challenges from Genomics and Molecular Biology; Michael Kearns: Computational game theory; Andrew McCallum: Information extraction from the world wide web; Sebastian Seung: Neural integrators; Yair Weiss, Jianbo Shi & Serge Belongie: Eigenvector methods for clustering and image segmentation SUBMISSIONS: The NIPS*2002 categories for paper submission are listed below. The subcategories are by no means exhaustive. Algorithms and Architectures: statistical learning algorithms, neural network architectures, kernel methods, graphical models, Gaussian processes, independent component analysis, model selection, active learning, combinatorial optimization. Applications: innovative applications or fielded systems that use machine learning, including time series, biological applications, text/web analysis, multimedia, robotics, or other intelligent systems. Cognitive Science/Artificial Intelligence: theoretical, computational, or experimental studies of perception, psychophysics, human or animal learning, memory, reasoning, problem solving, language, and neuropsychology. Emerging Technologies: analog and digital VLSI, neuromorphic engineering, computational sensors and actuators, microrobotics, bioMEMS, neural prostheses, photonics, molecular and quantum computing. Neuroscience: theoretical and experimental studies of encoding, decoding, processing, and transmission of information in biological neurons, including spike train generation, synaptic modulation, plasticity and adaptation, and network properties. Reinforcement Learning and Control: Markov decision processes, exploration, planning, navigation, game-playing, multi-agent coordination, computational models of classical and operant conditioning. Speech and Signal Processing: recognition, coding, synthesis, de-noising, source separation, auditory perception, psychoacoustics, temporal algorithms for signal processing such as Markov models, dynamical systems, recurrent networks. Theory: learning theory, information theory, statistical physics of learning, Bayesian methods, approximation bounds, online learning and dynamics, generalization and regularization. Visual Processing: 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 newly created demonstration track should consult the Web site below for more detailed instructions. REVIEW CRITERIA: All submitted papers will be thoroughly refereed on the basis of technical quality, significance, and clarity. Authors new to NIPS are particularly encouraged to submit their work. There will be an opportunity after the meeting to revise accepted manuscripts before submitting a final camera-ready copy for the proceedings. PAPER FORMAT: Submitted papers 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 8.25in by 5in rectangle. Submissions failing to follow these guidelines will not be considered. Authors are strongly encouraged to use the NIPS LaTeX style files obtainable from the web site listed below. SUBMISSION INSTRUCTIONS: NIPS accepts only electronic submissions. Full submission instructions will be available at the web site given below. NIPS accepts submissions in postscript and PDF format. The electronic submission process will begin on June 9, 2002. SUBMISSION DEADLINE: SUBMISSIONS MUST BE LOGGED BY MIDNIGHT JULY 1, 2002, PACIFIC DAYLIGHT TIME. The LaTeX style files, the electronic submission page, and other conference information are available on the web at http://nips.cc NIPS*2002 ORGANIZING COMMITTEE: General Chair, Sue Becker, McMaster University; Program Chair, Sebastian Thrun, Carnegie Mellon University; Publications Chair, Klaus Obermayer, TU Berlin; Tutorial Chair, Lawrence Saul, University of Pennsylvania; Workshops Co-chairs, Barak Pearlmutter, University of New Mexico, Robert Jacobs, University of Rochester; Demonstrations Co-chairs, Shihab Shamma, University of Maryland, Shih-Chii Liu, University and ETH Zurich; Publicity Chair, Zoubin Ghahramani, University College London; Volunteer Coordinator, Rajesh Rao, University of Washington; Treasurer, Bartlett Mel, University of Southern California; Web Masters, Alex Gray, Carnegie Mellon University, Guy Lebanon, Carnegie Mellon University; Contracts, Steve Hanson, Rutgers University. NIPS*2002 PROGRAM COMMITTEE: Sebastian Thrun, Carnegie Mellon University (chair); Peter Bartlett, BIOwulf Technologies and University of California, Berkeley; Gert Cauwenberghs, Johns Hopkins University; Geoffrey Gordon, Carnegie Mellon University; Daniel Lee, University of Pennsylvania; Marina Meila, University of Washington; Klaus-Robert Mueller, Fraunhofer FIRST ; Andrew Y. Ng, University of California, Berkeley; John Platt, Microsoft Research; Sam Roweis, University of Toronto; Eero Simoncelli, New York University; Joshua Tenenbaum, MIT; Chris Williams, University of Edinburgh; Richard Zemel, University of Toronto. PAPERS MUST BE SUBMITTED BY JULY 1, 2002 -- PLEASE POST -- From becker at nips.salk.edu Mon Apr 8 17:14:48 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:14:48 -0700 (PDT) Subject: NIPS*2002 call for workshop proposals Message-ID: _________________________________________________________________ Neural Information Processing Systems Natural and Synthetic NIPS*2002 Post-Conference Workshops December 13 and 14, 2002 Call for Workshop Proposals Whistler/Blackcomb Resort, BC, CANADA _________________________________________________________________ Following the regular program of the Neural Information Processing Systems 2002 conference in Vancouver, BC, Canada, workshops on various current topics in neural information processing will be held on December 13 and 14, 2002, in Whistler, BC, Canada. We invite researchers interested in chairing one of these workshops to submit workshop proposals. 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 particularly encouraged. Workshop topics include, but are not limited to, the following: Active Learning, Attention, Audition, Bayesian Analysis, Bayesian Networks, Benchmarking, Brain Imaging, Computational Complexity, Computational Molecular Biology, Control, Genetic Algorithms, Graphical Models, Hippocampus and Memory, Hybrid Supervised/Unsupervised Learning, Hybrid HMM/ANN Systems, Implementations, Independent Component Analysis, Mean-Field Methods, Markov Chain Monte-Carlo Methods, Music, Network Dynamics, Neural Coding, Neural Plasticity, On-Line Learning, Optimization, Recurrent Nets, Robot Learning, Rule Extraction, Self-Organization, Sensory Biophysics, Signal Processing, Spike Timing, Support Vectors, Speech, Time Series, Topological Maps, and Vision. Detailed descriptions of previous workshops may be found at http://www-2.cs.cmu.edu/Web/Groups/NIPS/NIPS2001/prevconf.html. There will be six hours of workshop meetings per day, split into morning and afternoon sessions, with free time in between 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 evening plenary sessions -- Writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. SUBMISSION INSTRUCTIONS Interested parties should submit a short proposal for a workshop of interest via email by August 9, 2002. Proposals should include 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 pure "mini-conference" format. An example format is: Tutorial lecture providing background and introducing terminology relevant to the topic. Two short lectures introducing different approaches, alternating with discussions after each lecture. 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% 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 motivate why the topic is of interest or controversial, why it should be discussed, and who the targeted group of participants is. It also should 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-workshop-proposal at cs.unm.edu. Please do not use attachments, Microsoft Word, postscript, html, or pdf files. Questions may be addressed to nips-workshop-admin at cs.unm.edu. Information about the main conference and the workshop program can be found at http://nips.cc/. Barak A. Pearlmutter, University of New Mexico Robert A. Jacobs, University of Rochester NIPS*2002 Workshops Co-Chairs PROPOSALS MUST BE RECEIVED BY AUGUST 9, 2002 --- Please Post --- From becker at nips.salk.edu Mon Apr 8 17:15:41 2002 From: becker at nips.salk.edu (Sue Becker) Date: Mon, 8 Apr 2002 14:15:41 -0700 (PDT) Subject: NIPS*2002 call for demonstrations Message-ID: *** New at NIPS: Demonstrations Track *** __________________________________________________________________ Neural Information Processing Systems Natural andSynthetic CALL FOR DEMONSTRATIONS Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada} __________________________________________________________________ For the first time, the Neural Information Processing Systems conference will include a separate track for demonstrations. The demonstrations will take place in parallel with the poster sessions at the NIPS*2002 conference. Example areas of interest for the demonstrations track include but are by no means limited to the following: Analog and digital VLSI Neuromorphic Engineering Computational sensors and actuators Robotics bioMEMS (microelectromechanical systems) Biomedical instrumentation Neural prostheses Photonics Real-time multimedia systems Large-scale neural emulators Software demonstrations of novel algorithms NIPS is an interdisciplinary conference, which attracts cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The demonstration track enables researchers to highlight scientific advances, systems, and technologies in ways that go beyond conventional poster presentations. It will provide a unique forum for demonstrating advanced technologies (hardware and software), and fostering the direct exchange of knowledge. We hope that this track will stimulate interactions between researchers from different fields (for example, roboticists and neuromorphic engineers) and encourage new colloboration between researchers in theoretical fields and those in more applied fields. Submissions accepted in the demonstrations track will be published on the NIPS web site, but will not appear in printed proceedings. However, submitting your work to the demonstration track by no means precludes the submission of a companion paper to the regular NIPS conference. In fact, joint submissions are very much encouraged. We also encourage authors submitting demonstrations to consider organizing a workshop at NIPS*2002. Note that the deadline for paper submissions is July 1, 2002, and for demonstration and workshop proposals the submission deadline is August 9, 2002. Please see http://nips.cc for further details. There will be a separate room for these demonstrations and participants will have access to power strips, tables and poster boards. VCRs and monitors will also be provided on request. Participants are responsible for ensuring that their demonstration is sufficiently portable; additional hardware beyond that specified above will not be provided by NIPS. SUBMISSION INSTRUCTIONS: All proposals for demonstrations will be reviewed by the Demonstrations Co-Chairs. Interested parties should submit a brief description of their proposed demonstration via email by August 9, 2002. Proposals should include a title, description of the device or system to be demonstrated, main results, novelty and significance of the work, any related publications, and estimated space requirements for the demonstration. Please include the name, address, email address, phone and fax numbers for all co-authors on the submmitted work, and indicate whether a related paper has also been submitted to NIPS*2002. Proposals should be emailed to shih at ini.phys.ethz.ch, and should be in plain ascii text, postscript or pdf. Questions may be addressed to shih at ini.phys.ethz.ch. Information about the main conference and the workshop program can be found at http://nips.cc Shihab Shamma Electrical Engineering Department & Institute for Systems Research University of Maryland and Shih-Chii Liu Institute of Neuroinformatics, University of Zurich & ETH Zurich NIPS*2002 Demonstrations Co-Chairs PROPOSALS MUST BE RECEIVED BY AUGUST 9, 2002 --- Please Post --- From rsun at cecs.missouri.edu Mon Apr 8 17:12:33 2002 From: rsun at cecs.missouri.edu (rsun@cecs.missouri.edu) Date: Mon, 8 Apr 2002 16:12:33 -0500 Subject: Cognitive Systems Research, Volume 3, Issue 1 Message-ID: <200204082112.g38LCXB08496@ari1.cecs.missouri.edu> A new issue is now available for * Cognitive Systems Research, Volume 3, Issue 1, Pages 1-119 (March 2002) =============================================================================== Cognitive Systems Research, Volume 3, Issue 1, Pages 1-119 (March 2002) TABLE OF CONTENTS Introduction to the special issue on computational cognitive modeling, Pages 1-3 Christian D. Schunn and Wayne D. Gray http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-3/1/711b118ed203aeda9a398ad63954e8dc A modular neural-network model of the basal ganglia's role in learning and selecting motor behaviours, Pages 5-13 Gianluca Baldassarre http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-1/1/35c9176c4afb2fb3a3ba329b618af22a Bootstrapping in miniature language acquisition, Pages 15-23 Rutvik Desai http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-4/1/747206abfb974b02bf41c156c86950bb Modeling icon search in ACT-R/PM, Pages 25-33 Michael D. Fleetwood and Michael D. Byrne http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-5/1/aa433104bb8019311d4f5630752c4b8a In search of templates, Pages 35-44 Fernand Gobet and Samuel Jackson http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-6/1/b8468b413f4ca0b9c2471e277a46095c An attractor network model of serial recall, Pages 45-55 Matt Jones and Thad A. Polk http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-7/1/b31a07d56e1992bf0131d5ffd6b0cf95 Intention superiority effect: A context-switching account, Pages 57-65 Christian Lebiere and Frank J. Lee http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-8/1/9c02678c6285705623482bff50cb8ca1 Modeling selective attention: Not just another model of Stroop (NJAMOS), Pages 67-76 Marsha C. Lovett http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-2/1/416e6f6b5973838ca291ccf14a772ec6 Extending task analytic models of graph-based reasoning: A cognitive model of problem solving with Cartesian graphs in ACT-R/PM, Pages 77-86 David Peebles and Peter C. -H. Cheng http://www.sciencedirect.com/science/article/B6W6C-44D4KP4-3/1/c0b76a0b1f47140f3067fddc1079731d The role of computational modeling in understanding hemispheric interactions and specialization, Pages 87-94 James A. Reggia and Reiner Schulz http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-1/1/319c267472df4cc3919c2103d3d425ba Predicting the effects of cellular-phone dialing on driver performance, Pages 95-102 Dario D. Salvucci and Kristen L. Macuga http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-9/1/57923a5f59840882407bfd66fb61a599 A model of individual differences in skill acquisition in the Kanfer-Ackerman air traffic control task, Pages 103-112 Niels A. Taatgen http://www.sciencedirect.com/science/article/B6W6C-44GHSVC-2/1/f2a7daf9433c6074808d7fd946c19afc An explanation of the length effect for rotated words, Pages 113-119 Carol Whitney http://www.sciencedirect.com/science/article/B6W6C-44HTMB3-1/1/452c70ac314fb5e174c1a994a7103278 =============================================================================== See the following journal Web pages for subscription information: http://www.elsevier.com/locate/cogsys http://www.cecs.missouri.edu/~rsun/journal.html =========================================================================== Prof. Ron Sun http://www.cecs.missouri.edu/~rsun CECS Department phone: (573) 884-7662 University of Missouri-Columbia fax: (573) 882 8318 201 Engineering Building West Columbia, MO 65211-2060 email: rsun at cecs.missouri.edu http://www.cecs.missouri.edu/~rsun http://www.cecs.missouri.edu/~rsun/journal.html http://www.elsevier.com/locate/cogsys =========================================================================== From J.A.Bullinaria at cs.bham.ac.uk Tue Apr 9 08:01:02 2002 From: J.A.Bullinaria at cs.bham.ac.uk (John A Bullinaria) Date: Tue, 9 Apr 2002 13:01:02 +0100 (BST) Subject: Lectureships / Seinor Lectureships at Birmingham, UK Message-ID: Dear Connectionists, The University of Birmingham is looking for three additional Lecturers / Seinor Lecturers in Computer Science, Software Engineering or Artificial Intelligence. The specific areas of research are open. Birmingham already has a strong research group in natural computation, and we would warmly encourage outstanding applicants from the broad field of nature inspired computation (including neural networks, evolutionary computation, etc.). There are currently eight members of academic staff working on natural computation and related topics: Dr. John Bullinaria (Neural Networks, Evolutionary Computation, Cog.Sci.) Dr. Ke Chen (Neural Networks, Pattern Recognition, Machine Perception) Dr. Aniko Ekart (Genetic Programming, AI, Machine Learning) Dr. Jun He (Evolutionary Computation, Artificial Immune Systems) Dr. Julian Miller (Evolutionary Computation, Machine Learning) Dr. Jon Rowe (Evolutionary Computation, AI) Dr. Thorsten Schnier (Evolutionary Computation, Engineering Design) Prof. Xin Yao (Evolutionary Computation, NNs, Nature Inspired Comp.) Other staff members also working in these areas include Prof. Aaron Sloman (evolvable architectures of mind, co-evolution, interacting niches), Dr. Jeremy Wyatt (evolutionary robotics, reinforcement learning), and Dr Ela Claridge (evolutionary image processing). The official job advert is attached below. Best regards, John Bullinaria. ========================================================================== THE UNIVERSITY OF BIRMINGHAM School of Computer Science (http://www.cs.bham.ac.uk/) NEW ACADEMIC POSTS AVAIABLE IN COMPUTER SCIENCE, SOFTWARE ENGINEERING OR ARTIFICIAL INTELLIGENCE. Applications are invited for three open-ended (i.e. not temporary) appointments, at the level of Lecturer or Senior Lecturer, in Computer Science, Software Engineering or Artificial Intelligence. One post is available to start from 1 September 2002 (or later), the remaining two will be available from 01 January 2003 (or later): exact start dates are negotiable. The appointees will be expected to contribute stongly to research, teaching and administration within the School. We would welcome applicants who wish to lead the School into new research or teaching areas as well as those who wish to add to the School's existing strengths. The School also encourages industrial collaboration and consultancy. The School obtained a grade of 5 in the 2001 national Research Assessment Exercise. This shows international recognition of the excellence of the School's research. The School is now starting a new phase of expansion over several years which is planned to include the appointment of a new professor (equivalent to full professor in the USA). Several additional posts will be advertised later, with start dates in 2003. Applicants should have an internationally excellent research background in Computer Science, Software Engineering or Artificial Intelligence, as evidenced by publications in leading international journals or conference proceedings in these fields. Applicants should have teaching experience in one of the fields, and should have, or expect soon to have, a PhD in one of the fields or an appropriate, closely related discipline. Applicants for a Senior Lecturership should already meet our criteria for promotion to that level. For these criteria, please contact Mr David King, tel: (+44)(0)121 414 3711; Email: D.J.King at cs.bham.ac.uk. Starting salary for Lecturers on scale £20,470 - £32,537 a year depending on experience and qualifications. Starting salary for Senior Lecturers on scale £34,158 - £38,603 year depending on experience and qualifications. (The university pays additional pension contributions.) Informal enquiries to: Professor Aaron Sloman Tel: (+44) (0)121 414 3711 or email: A.Sloman at cs.bham.ac.uk If possible please avoid sending him Word files: use plain text or html, or postscript or PDF. Please state in any application whether you are interested in a Lectureship or Senior Lectureship. Application forms returnable by 25 April 2002 (late applications may be considered). Further particulars including instructions on how to apply, and pointers to online application forms, can be found at: http://www.cs.bham.ac.uk/news/jobs/Lectureship2002.html Please quote reference S35803/02. Further information about the School and its Research and teaching can be found here: http://www.cs.bham.ac.uk/ http://www.cs.bham.ac.uk/research/ http://www.cs.bham.ac.uk/news/ http://www.cs.bham.ac.uk/study/ Working towards equal opportunities. ========================================================================== From ken at phy.ucsf.edu Tue Apr 9 18:57:05 2002 From: ken at phy.ucsf.edu (Ken Miller) Date: Tue, 9 Apr 2002 15:57:05 -0700 Subject: Paper available: reverse suture and orientation map development Message-ID: <15539.29121.257475.147355@coltrane.ucsf.edu> The following paper is now available from ftp://ftp.keck.ucsf.edu/pub/ken/miller-erwin01.ps (postscript, 4 MB) or ftp://ftp.keck.ucsf.edu/pub/ken/miller-erwin01.ps.gz (gzipped ps, 1.2 MB) or from http://www.keck.ucsf.edu/~ken (click on 'publications', then on 'models of neural development') "Effects of Monocular Deprivation and Reverse Suture On Orientation Maps Can Be Explained By Activity-Instructed Development of Geniculocortical Connections" by K.D. Miller and E. Erwin This is a final draft of a paper that has now appeared as Visual Neuroscience 18:821-834 (2001). Abstract: Mature visual cortex shows a single, binocularly matched orientation map. This matching develops without visual experience. It persists despite early monocular deprivation that largely eliminates one eye's map, followed by reverse suture (deprivation of the previously open eye and opening of the previously deprived eye), even though the two eyes lack common visual experience in this case. These results have been interpreted to suggest that the structure of orientation maps either is innately predetermined or, if it arises through self-organization, is determined by external cues such as boundary conditions or a ``scaffolding'' of horizontal connections. We show, to the contrary, that these results are the expected outcomes if orientation maps develop through activity-instructed, correlation-based development of the geniculocortical connections without additional cues. A weak, binocularly correlated orientation map is known to exist before deprivation onset; we previously showed how this can arise through activity-instructed development. Now we show that this initial correlation between the two eyes' maps can persist or increase despite deprivation sufficient to cause massive loss of the deprived eye's geniculocortical synaptic strength, followed by reverse suture. Given sufficient early correlated map development, each map's fate is ``dynamically committed'': the two eyes' maps will converge upon a common outcome, even if developing independently. This dynamic fate commitment is retained even after severe deprivation. Ken Kenneth D. Miller telephone: (415) 476-8217 Associate Professor fax: (415) 476-4929 Dept. of Physiology, UCSF internet: ken at phy.ucsf.edu 513 Parnassus www: http://www.keck.ucsf.edu/~ken San Francisco, CA 94143-0444 From castellano at di.uniba.it Thu Apr 11 06:37:59 2002 From: castellano at di.uniba.it (Giovanna Castellano) Date: Thu, 11 Apr 2002 12:37:59 +0200 Subject: call for contribution at invited session in KES2002 Message-ID: <00af01c1e144$f31eeb00$c9bbccc1@castellano> As you may aware, the coming KES Conference will be held in September this year in Italy (http://www.bton.ac.uk/kes/kes2002/) I am currently organizing a special session for the conference in the area of Knowledge-based neurocomputing systems. I would like to invite you to submit papers to this session. The original deadline for submissions is 1 April. However, with the understanding that many of us may be busy and unable to meet this deadline, I have negotiated a further extension of the deadline to 20 April. So if you are interested in contributing to this session please let me know at your earliest convenience, sending ASAP your intention to submit a paper to me (castellano at di.uniba.it)l. You can find below the cfp containing information for paper submission. If you have any questions in that regard please feel free to contact me. Dr. Giovanna CASTELLANO Assistant Professor, PhD in Computer Science Computer Science Department University of Bari Via Orabona, 4 70125 Bari, ITALY phone: +39 080 5442456 fax : +39 080 5443196 e-mail: castellano at di.uniba.it URL: www.di.uniba.it/people/castellano.htm ================================================================ KES'2002 Sixth International Conference on Knowledge-Based Intelligent Information Engineering Systems 16, 17 & 18 September 2002 Podere d'Ombriano, Crema, Italy Special Session: Knowledge-Based Neurocomputing Systems Call for papers This special session aims at presenting and discussing Knowledge-Based Neurocomputing Systems, which concern the use and the explicit representation of problem-specific knowledge within the neurocomputing paradigm. All areas of simulations where KBN systems can be applied will be considered, giving a special focus to applications in medicine. Topics of interest include (but are not limited to): a.. Hybrid architectures for neural networks b.. Fuzzy and neurofuzzy techniques c.. Methods of rule extraction/discovery d.. Clustering and classification e.. Pattern recognition f.. Predictive models g.. Control / Optimization systems h.. Medical and diagnostic systems Submissions Papers should be submitted in electronic format (preferably in PDF or PostScript) to castellano at di.uniba.it, using the subject-line "KES02 Special Session submission". Papers must be formatted as detailed in IOS Instructions for the Preparation of a Camera-Ready Manuscript (see the KES 2002 homepage for details). The maximum length of papers is five pages (when prepared according to the IOS instructions). Longer papers will be subject to an additional page charge. Please note that final versions of accepted papers must be submitted in hard copy. Submitted papers have to be unpublished, containing new and original results. Submission implies the willingness of at least one of the authors to register and present the paper. The session papers will be published in the Conference Proceedings (IOS Press, Amsterdam, the Netherlands). Extended versions of selected papers will be considered for publication in the KES Journal (International Journal of Knowledge-Based Intelligent Engineering Systems). Important Dates Submission deadline: April 20, 2002 Notification of acceptance: May 15, 2002 Deadline for camera-ready papers: June 1, 2002 Session Organizer and Chair Dr. Giovanna Castellano Computer Science Department University of Bari Via Orabona, 4 70125 Bari, ITALY Tel: +39 080 5442456 Fax: +39 080 5443196 email: castellano at di.uniba.it From shastri at ICSI.Berkeley.EDU Thu Apr 11 15:04:19 2002 From: shastri at ICSI.Berkeley.EDU (Lokendra Shastri) Date: Thu, 11 Apr 2002 12:04:19 PDT Subject: Paper on computational abstraction of long-term potentiation Message-ID: <200204111904.MAA22112@dill.ICSI.Berkeley.EDU> Dear Connectionists, The following article may be of interest to you: "A Computationally Efficient Abstraction of Long-term Potentiation" Neurocomputing (In Press). The final draft is available at: http://www.icsi.berkeley.edu/~shastri/psfiles/ShastriNCltp02.pdf Best wishes, -- Lokendra Shastri http://www.icsi.berkeley.edu/~shastri Phone: (510) 666-2910 FAX: (510) 666-2956 --------------------------------------------------------------------- A Computationally Efficient Abstraction of Long-term Potentiation Lokendra Shastri ICSI, Berkeley, CA. Key words: Long-term potentiation; hippocampus; binding detection; dentate gyrus; episodic memory Abstract A computational abstraction of long-term potentiation (LTP) is proposed. The abstraction captures key temporal and cooperative properties of LTP, and also lends itself to rapid computation. The abstraction is used to simulate the recruitment of binding-detector cells in the dentate gyrus (DG). The simulation results are used to validate a quantitative analysis of binding-detector cell recruitment in DG. The analysis shows that (i) a large number of binding-detector cells would be recruited in response to entorhinal cortex activity and (ii) cells recruited for distinct bindings would exhibit very low cross-talk. These results help in understanding the neural basis of episodic memory. From tbl at cin.ufpe.br Fri Apr 12 10:21:25 2002 From: tbl at cin.ufpe.br (Teresa Bernarda Ludermir) Date: Fri, 12 Apr 2002 11:21:25 -0300 Subject: Deadline Extension - VII BraziDeadline Extension - VII BrazilianDeadline Extension - VII Brazilian Symposium on Neural Networks (SBRN 2002) Message-ID: <3CB6ED65.C12EBAC7@cin.ufpe.br> Because of numerous requests for extensions, the deadline for submission of papers to the VII Brazilian Symposium on Neural Networks (SBRN 2002) is now April 22, 2002. This symposium will be held in Porto de Galinhas, Recife, Brazil from November 11-14, 2002, and is chaired by Professor Teresa Ludermir from the Federal University of Pernambuco in Brazil. All information regarding this symposium can be found on our website at http://www.cin.ufpe.br/~sbiarn02/sbrn02.html Please refer to this website for formatting and submission requirements. If you have any questions about the symposium, please feel free to contact us at sbrn2002 at cin.ufpe.br. Sincerely, Marcilio C. P. de Souto Program Chair - SBRN 2002 -- Teresa B. Ludermir http://www.cin.ufpe.br/~tbl Centro de Informatica - UFPE Phone: [+55 81] 3271-8430 Caixa Postal 7851 x: 4308 CEP 50732-970 Recife PE Brazil From sfr at unipg.it Fri Apr 12 09:11:58 2002 From: sfr at unipg.it (Simone G.O. Fiori (Pg)) Date: Fri, 12 Apr 2002 15:11:58 +0200 (DFT) Subject: Announcement of Tutorial on Neural Networks. Message-ID: <1018617118.3cb6dd1e8b8bb@webmail.unipg.it> Dear Colleagues, I am pleased to announce that during the Tenth Biennial IEEE Conference on Electromagnetic Field Computation (CEFC), to be held in Perugia (Italy) on June 15-19, 2002, a tutorial about the application of artificial neural networks to electromagnetic compatibility will be given. The tutorial is intended for young investigators interested in the roots of neural networks theory as well as for senior researchers in the neural field interested in the latest issues about networks' application to electro- magnetics. The tutorial program includes therefore topics such as the basic theory of multilayer perceptrons and their applications as well as emerging neural techniques such as independent component analysis. The title and time-scheduling of the full-day tutorial are: Neural Network Modeling of Electromagnetic Interaction Phenomena (Instructor: Dr. Simone Fiori, Perugia University) Saturday, June 15, 2002 (Time-scheduling: 9,00-13,00 and 15,00-18,00) The cost* for the attendances is: EUR 100/full registration including course notes, EUR 65/student registration including course notes. * Please note that, in order to attend the tutorial, it is not mandatory to subscribe to the conference. A short description of the conference is attached below for your convenience. For more information please visit the conference web-page at: http://www.unipg.it/cefc2002/ or contact the instructor (S. Fiori) at the email address sfr at unipg.it or the conference general chair (E. Cardelli) at the email address cefc2002 at unipg.it. Sincerely, S. Fiori Faculty of Engineering, Perugia University (Italy) ================================================================== The Tenth Biennial IEEE Conf. on Electromagnetic Field Computation: The Tenth Biennial IEEE Conference on Electromagnetic Field Computation (CEFC) will be held at Perugia during June 15-19, 2002. Perugia is an ancient town founded in the prehistoric epoch, located in the Umbria region, also called "the green heart of Italy" for its rich and extensive vegetation. Perugia is a very popular tourist destination, known for its important monuments, its folklore, its rich cuisine and its quiet and friendly people. The last Conference was held in Milwaukee, Wisconsin, USA in 2000. The aims of the IEEE CEFC are to present the latest developments in modelling and simulation methodologies for the analysis of electromagnetic fields and wave interactions, with the application emphasis being on the computer-aided design of low and high frequency devices, components and systems. Scientists and engineers worldwide will present and discuss original contributions in the areas of static and quasi-static fields; wave propagation; material modelling; coupled problems, optimisation; numerical techniques, software methodology; applications of electromagnetic CAD to electrical/electronic devices, components and systems prototyping. The Conference will feature oral and poster presentations. From l.s.smith at cs.stir.ac.uk Fri Apr 12 08:16:33 2002 From: l.s.smith at cs.stir.ac.uk (Leslie Smith) Date: Fri, 12 Apr 2002 13:16:33 +0100 Subject: 2 research assistant posts at Stirling University (Scotland) Message-ID: <3CB6D021.BD8FDA05@cs.stir.ac.uk> University of Stirling Department of Computing Science and Mathematics Research Assistant (2 posts) UK?17,626 - UK?19,681 The University of Stirling (jointly with the Universities of Glasgow and Oxford) has recently been awarded two major EPSRC research grants on communication between real neurons and electronics, and on spiking neural systems. Consequently, it has the following opportunities: Post 1: Communication between real neurons and electronics: Signalling between cultured neurons and electronics Three year Research Assistant post concerned primarily with the development of signals to be applied to neural systems, and with the interpretation of spikes emitted by these neural systems. Reference No: 6156/920 Post 2: Spiking neural systems: Spiking in low-level auditory processing Three year Research Assistant post developing neuromorphic spike-based algorithms for implementation in silicon (at Oxford and Edinburgh) for solving problems in early auditory processing. Reference No: 6157/927 For informal discussion contact Prof. Leslie Smith, lss at cs.stir.ac.uk tel (44) 1786 467435. For more information about the Department please see www.cs.stir.ac.uk. There will also be related posts at Edinburgh: see www.ee.ed.ac.uk/ISG/ Further particulars are available from the Personnel Office, University of Stirling, Stirling, FK9 4LA. Tel: (01786) 467028, Fax: (01786) 466155, e-mail personnel at stir.ac.uk Closing date for applications: Monday 6 May 2002. www.personnel at stir.ac.uk AN EQUAL OPPORTUNITIES EMPLOYER -- Professor Leslie S. Smith, Head of Department, Dept of Computing Science and Mathematics, University of Stirling, Stirling FK9 4LA, Scotland l.s.smith at cs.stir.ac.uk Tel (44) 1786 467435 Fax (44) 1786 464551 www http://www.cs.stir.ac.uk/~lss/ From D.Willshaw at anc.ed.ac.uk Mon Apr 15 07:05:15 2002 From: D.Willshaw at anc.ed.ac.uk (David Willshaw) Date: Mon, 15 Apr 2002 12:05:15 +0100 (BST) Subject: Edinburgh Doctoral Training in Neuroinformatics Message-ID: <15546.46059.403874.353053@gargle.gargle.HOWL> ------------------------------------------------------------------------ 4 YEAR DOCTORAL TRAINING IN NEUROINFORMATICS http://www.anc.ed.ac.uk/neuroinformatics first application deadline: May 1st The University of Edinburgh has just established a Doctoral Training Centre in Neuroinformatics. The programme is made up of 3 themes: 1) Computational and Cognitive Neurocience - analytical and computational modelling of information-processing in the nervous system; 2) Neuromorphic Engineering and Neural Computation - Artificial sensor perception and analysis, neuromorphic modelling, spiking computation and mixed-mode VLSI, evolutionary optimisation of physical systems, neurally inspired algorithms; 3) Simulation, Analysis, Visualisation and Data Handling - software systems and computational techniques for neuroscience and neural engineering The 4 year programme in Neuroinformatics consists of an introductory year of training in neuroscience and this broad range of application areas culminating in an MSc by Research in Neuroinformatics. This is followed by 3 years of Ph.D. study, building on the research project undertaken in the first year. Applications for 4 year studentships for entry in October 2002 are invited from EU students (see website for full eligibility). The stipend is set in the region of 13,000 pounds per annum. Full fees are paid. Note particularly that this studentship scheme is open to all students within the EU and is not restricted to applicants from the UK. Students with a strong background in computer science, electronics, mathematics, physics or engineering are particularly welcome, and we will consider those with other backgrounds. We will be accepting an average of 10 students a year into this programme, which will be located within the Institute for Adaptive and Neural Computation in the Division of Informatics, the UK's largest highest-quality academic computer-science grouping. Additional faculty are being recruited to support the Doctoral Training Centre and to grow Neuroinformatics at Edinburgh. Edinburgh has strong programmes of research in all of the areas listed above and leads the UK in integrating these into a coherent programme in Neuroinformatics. For full application details, and further information of areas of study at the Centre, consult the website: http://www.anc.ed.ac.uk/neuroinformatics or send email to neuroinformatics-phd at anc.ed.ac.uk Applications received by MAY 1 2002 will receive priority treatment. From bengio at idiap.ch Mon Apr 15 08:21:16 2002 From: bengio at idiap.ch (Samy Bengio) Date: Mon, 15 Apr 2002 14:21:16 +0200 (CEST) Subject: open postdoc positions in Machine Learning in Switzerland Message-ID: Open positions for Postdoctoral candidates in Machine Learning The Dalle Molle Institute for Perceptual Artificial Intelligence (IDIAP, http://www.idiap.ch) seeks qualified applicants for Postdoc positions in its Machine Learning group for research projects related to multimodal integration of asynchronous streams of information. The ideal candidate should have strong background in statistical learning theory, mixture models, and more specifically hidden Markov models and related sequence modeling tools. All applicants should be familiar with C/C++ programming under a Unix environment. Although IDIAP is located in the French part of Switzerland, English is the main working language at IDIAP. Free English and French courses are also provided. IDIAP is located in the town of Martigny (http://www.martigny.ch) in Valais, a scenic region in the south of Switzerland, surrounded by the highest mountains of Europe, and offering exciting recreational activities, including hiking, climbing and skiing, as well as varied cultural activities. It is within close proximity to Montreux (Jazz Festival) and Lausanne (EPFL, http://www.epfl.ch). Candidates should send their detailed CV to Dr. Samy Bengio Research Director. Machine Learning Group Leader. IDIAP, CP 592, rue du Simplon 4, 1920 Martigny, Switzerland. tel: +41 27 721 77 39, fax: +41 27 721 77 12. mailto:bengio at idiap.ch, http://www.idiap.ch/~bengio ----- Samy Bengio Research Director. Machine Learning Group Leader. IDIAP, CP 592, rue du Simplon 4, 1920 Martigny, Switzerland. tel: +41 27 721 77 39, fax: +41 27 721 77 12. mailto:bengio at idiap.ch, http://www.idiap.ch/~bengio From hamilton at may.ie Mon Apr 15 17:20:26 2002 From: hamilton at may.ie (Hamilton Institute) Date: Mon, 15 Apr 2002 22:20:26 +0100 Subject: Statisical Machine Learning PhD Position - Hamilton Institute/DaimlerChrysler Research Message-ID: <007b01c1e4c3$5dc7b7d0$06c09d95@pwhm6> JOINT PHD POSITION, HAMILTON INSTITUTE/DAIMLERCHRYSLER RESEARCH The Hamilton Institute invites applications for a PhD position in the general area of advanced automotive control. The project will involve utilising and developing concepts in some or all of the following areas: decentralised and interacting control systems; hybrid dynamical systems; statistical machine learning. This post is part of a collaboration with DiamlerChrysler Research (Stuttgart), the Center for Systems Science (Yale University) and the Hamilton Institute (NUI Maynooth). Candidates will be primarily based at the Hamilton Institute and at DaimlerChrysler Research. Successful candidates are likely to have demonstrated an outstanding level of academic achievement. These posts offer an exciting opportunity to tackle fundamental research problems within a stimulating multi-disciplinary research environment with state of the art facilities and strong links to the international research community. Applicants should send a C.V., including details of three referees to hamilton at may.ie. For further details, please visit www.hamilton.may.ie. The closing date for applications is the 31st May 2002. From Zoubin at gatsby.ucl.ac.uk Wed Apr 17 20:43:13 2002 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 18 Apr 2002 01:43:13 +0100 (BST) Subject: NIPS 2001 Proceedings Now Available Online Message-ID: <200204180043.BAA20546@crick.gatsby.ucl.ac.uk> The full proceedings of the 2001 Neural Information Processing Systems Conference are now available online at: http://www.nips.cc/2001papers These will be published in Advances in Neural Information Processing Systems 14. T. G. Dietterich, S. Becker, and Z. Ghahramani (eds.) MIT Press, Cambridge, MA, 2002. Please note that the FULL TEXT of the proceedings is searchable if you have the djvu-plugin. We hope you enjoy this resource. Zoubin Ghahramani NIPS 2001 Publications Chair From wahba at stat.wisc.edu Wed Apr 17 17:52:16 2002 From: wahba at stat.wisc.edu (Grace Wahba) Date: Wed, 17 Apr 2002 16:52:16 -0500 (CDT) Subject: The Multicategory Support Vector Machine Message-ID: <200204172152.QAA03061@juno.stat.wisc.edu> I'm pleased to announce "Classification of Multiple Cancer Types by Multicategory Support Vector Machines Using Gene Expression Data" by Yoonkyung Lee and Cheol-Koo Lee, UW-Madison Statistics Dept TR 1051 (2002). This paper (and related papers) are accessible via the home pages: http://www.stat.wisc.edu/~yklee or http://www.stat.wisc.edu/~wahba Abstract: Monitoring gene expression profiles is a novel approach in cancer diagnosis. Several studies showed that prediction of cancer types using gene expression data is promising and very informative. The Support Vector Machine (SVM) is one of the classification methods successfully applied to the cancer diagnosis problems using gene expression data. However, its extension to more than two classes was not obvious, which might impose limitations in its application to multiple tumor types. In this paper, we analyze a couple of published multiple cancer types data sets by the multicategory SVM, which is a recently proposed extension of the binary SVM. ...................... ....................... The Multicategory Support Vector Machine was proposed in "Multicategory Support Vector Machines", Yoonkyung Lee, Yi Lin and Grace Wahba UW-Madison Statistics Dept TR 1043 (2001), also accessible as above. ......................... Multicategory `Soft' classification was proposed in "Smoothing Spline Analysis of Variance for Polychotomous Response Data" Xiwu Lin, UW-Madison Statistics Dept TR 1003 (1998), available via the wahba home pg. It can be argued that the SVM can be considered a `hard' classifier and while the penalized likelihood estimate (as in TR 1003)is a `soft' classifier, and it can be argued that the the SVM is more appropriate where the attribute data is relatively sparse and the category overlap is relatively small while the soft classifier may be more appropriate where the attribute data is relatively dense and the category overlap is more substantial. From ole.jensen at fcdonders.kun.nl Wed Apr 17 08:40:04 2002 From: ole.jensen at fcdonders.kun.nl (Ole Jensen) Date: Wed, 17 Apr 2002 14:40:04 +0200 Subject: paper announcement : evoked gamma Message-ID: <3CBD6D24.6070604@fcdonders.kun.nl> Dear Connectionists, I would to draw your attention to a paper recently published in Neuroimage: Jensen O, Hari R, Kaila K. (2002) Visually evoked gamma responses in the human brain are enhanced during voluntary hyperventilation. Neuroimage 15:575-586. The paper is a combined MEG and theoretical study on visually evoked gamma responses. By analysing a network of Hodgkin-Huxley type neurons we aim to account for properties of the visually evoked gamma response. Please contact me for reprints or visit my website: http://baserv.uci.kun.nl/~olejen Best regards, Ole Jensen ==================================================================== Ole Jensen, Ph.D. F.C. Donders Centre for Cognitive Neuroimaging P.O. Box 9101 NL-6500 HB Nijmegen The Netherlands Office : +31 24 36 10884 Mobile : +31 62 18 57495 Fax : +31 24 36 10652 e-mail : ole.jensen at fcdonders.kun.nl URL : http://baserv.uci.kun.nl/~olejen http://www.fcdonders.kun.nl From marcus at idsia.ch Thu Apr 18 08:14:48 2002 From: marcus at idsia.ch (Marcus Hutter) Date: Thu, 18 Apr 2002 14:14:48 +0200 Subject: New Paper on Mutual Information Message-ID: <00ad01c1e6d2$a3094960$0fbfb0c3@ho> Dear Colleagues, I would like to announce a new paper on Mutual Information beyond the point/frequency estimate, which today appeared in the online NIPS-Proceedings http://www-2.cs.cmu.edu/~nips/2001papers/ since it is possibly of general interest. Best wishes Marcus Hutter =========================================================================== "Distribution of Mutual Information" The mutual information of two random variables i and j with joint probabilities t_ij is commonly used in learning Bayesian nets as well as in many other fields. The chances t_ij are usually estimated by the empirical sampling frequency n_ij/n leading to a point estimate I(n_ij/n) for the mutual information. To answer questions like ``is I(n_ij/n) consistent with zero?'' or ``what is the probability that the true mutual information is much larger than the point estimate?'' one has to go beyond the point estimate. In the Bayesian framework one can answer these questions by utilizing a (second order) prior distribution p(t) comprising prior information about t. From the prior p(t) one can compute the posterior p(t|n), from which the distribution p(I|n) of the mutual information can be calculated. We derive reliable and quickly computable approximations for p(I|n). We concentrate on the mean, variance, skewness, and kurtosis, and non-informative priors. For the mean we also give an exact expression. Numerical issues and the range of validity are discussed. Downloadable also at: http://www.idsia.ch/~marcus/ai/xentropy.htm in various formats. =========================================================================== Dr. Marcus Hutter, IDSIA Istituto Dalle Molle di Studi sull'Intelligenza Artificiale Galleria 2 CH-6928 Manno(Lugano) - Switzerland Phone: +41-91-6108668 Fax: +41-91-6108661 E-mail marcus at idsia.ch http://www.idsia.ch/~marcus From bengioy at IRO.UMontreal.CA Thu Apr 18 09:06:11 2002 From: bengioy at IRO.UMontreal.CA (Yoshua Bengio) Date: Thu, 18 Apr 2002 09:06:11 -0400 Subject: PhD + post-doc positions in neural network motor models Message-ID: <20020418090611.E24965@einstein.iro.umontreal.ca> ***************************************************************** Visuomotor Control and Learning in Large-Scale Biological and Artificial Neural Networks Positions are available at the graduate and postdoctoral level to work with a team of researchers who are pursuing a series of collaborative behavioral, neurophysiological and computational studies of the neural mechanisms of visuomotor control and motor skill acquisition in the human, monkey and cat, and their application to the design of autonomous adaptive neural networks. Team members include John Kalaska, Yoshua Bengio, Trevor Drew, Paul Cisek and Stephen Scott. Physiological projects are studying the functional organization and learning mechanisms in mammalian motor systems. Modelling and theoretical tools are being used to develop and evaluate the performance of different computational architectures and learning algorithms for adaptive task decomposition in large-scale biological and artificial neural networks that are inspired by the physiological studies, and to assess their potential applications in a wide range of technological domains. Candidates have the opportunity to collaborate with several team members. Candidates must have a strong background in one or more of the following areas: learning mechanisms in artificial or biological neural networks, computer science, control theory, statistical analysis, computational neuroscience, systems neuroscience, motor control, biomechanics, psychophysics. Positions are available immediately, but start times are negotiable. Open to all nationalities. Graduate-level candidates will apply for a PhD at one of three departments: Physiology at U. Montreal, Montreal, Quebec (J. Kalaska, T. Drew, P. Cisek), Computer Science at U. Montreal (Y. Bengio), or Anatomy at Queens University, Kingston, Ontario (S. Scott). Competitive stipends. Funding available for periods of up to 3-5 years. Send CV, statement of research interests, and 2 letters of reference to John Kalaska, Dpartement de Physiologie, Pavillon Paul-G-Desmarais, Universit de Montral, 2900 Edouard-Montpetit, Montral Qubec, Canada, H3T 1J4 (telephone: 514-343-6349; fax: 514-343-6113; e-mail: kalaskaj at physio.umontreal.ca). ***************************************************************** -- Yoshua Bengio Associate professor / Professeur agrg Canada Research Chair in Statistical Learning Algorithms / titulaire de la chaire de recherche du Canada en algorithmes d'apprentissage statistique Dpartement d'Informatique et Recherche Oprationnelle Universit de Montral, adresse postale: C.P. 6128 Succ. Centre-Ville, Montral, Qubec, Canada H3C 3J7 adresse civique: 2920 Chemin de la Tour, Montral, Qubec, Canada H3T 1J8, #2194 Tel: 514-343-6804. Fax: 514-343-5834. Bureau 3339. http://www.iro.umontreal.ca/~bengioy http://www.iro.umontreal.ca/~lisa From dirk at bioss.ac.uk Fri Apr 19 04:10:56 2002 From: dirk at bioss.ac.uk (Dirk Husmeier) Date: Fri, 19 Apr 2002 09:10:56 +0100 Subject: PhD position in Bioinformatics Message-ID: <3CBFD110.B7E87015@bioss.ac.uk> A three-year PhD studentship is available at Biomathematics & Statistics Scotland (BioSS) for a highly motivated candidate to work on the application of machine learning techniques and graphical models in bioinformatics with a focus on phylogenetics and/or gene expression data analysis. The work will be carried out at the science campus of EdinburghUniversity, and is available immediately. Applicants should have, or shortly expect to obtain, a first or upper second class degree in mathematics, physics, statistics, mathematical biology or a related discipline, and strong programming skills in C/C++ and MATLAB. A familiarity with multivariate statistics, Bayesian methods, probabilistic graphical models, and numerical optimization (gradient ascent, EM algorithm, simulated annealing etc.) would be useful. Awards include full university fees (for EU nationals) and a maintenance allowance of =A310,000 per year. Please send a CV and the names and addresses of two referees to: Dirk Husmeier Biomathematics & Statistics Scotland By email: dirk at bioss.ac.uk By surface mail, until 30 April: SCRI, Inergowrie DUNDEE, DD2 5DA United Kingdom By surface mail, after 1 May: JCMB, The King's Buildings, EDINBURGH, EH9 3J United Kingdom All applications will be considered until the position has been filled. An outline of two possible projects is available at: http://www.bioss.sari.ac.uk/~dirk/temp/PhD_project.html -- ---------------------------------------------- Dirk Husmeier Biomathematics and Statistics Scotland (BioSS) SCRI, Invergowrie, Dundee DD2 5DA, United Kingdom http://www.bioss.ac.uk/~dirk/ From rens at science.uva.nl Fri Apr 19 06:48:39 2002 From: rens at science.uva.nl (Rens Bod) Date: Fri, 19 Apr 2002 12:48:39 +0200 (MET DST) Subject: Memory-based Models of Music Message-ID: Hi, The following paper-in-press may be of interest to the readers of this list: R. Bod. "Memory-Based Models of Melodic Analysis: Challenging the Gestalt Principles". Journal of New Music Research 30(3). See http://turing.wins.uva.nl/~rens/jnmr01.pdf Best wishes, Rens Bod (http://turing.wins.uva.nl/~rens) From dayan at gatsby.ucl.ac.uk Mon Apr 22 06:23:01 2002 From: dayan at gatsby.ucl.ac.uk (Peter Dayan) Date: Mon, 22 Apr 2002 11:23:01 +0100 Subject: PhD studentship with Wolfram Schultz and Peter Dayan Message-ID: <20020422112301.A3358@flies.gatsby.ucl.ac.uk> PhD Studentship Neuroscience Applications are invited for a research studentship on empirical and theoretical study of the neural basis of reinforcement learning, tenable as part of a collaboration between Wolfram Schultz, at the Department of Anatomy, University of Cambridge, and Peter Dayan, at the Gatsby Computational Neuroscience Unit, UCL. The Cambridge lab has long-standing experience in the recording and analysis of the activities of dopamine cells and their targets; the Gatsby Unit is one of the largest labs in the world devoted to computational neuroscience. Applicants should have a strong analytical background, a keen interest in neuroscience, and a good honours degree in a relevant subject. The post will likely be based in Cambridge, though substantial interaction with the Gatsby Unit in London (90 minutes away) will be required. Applications should include a CV, a statement of research interests, and a list of three referees, and should be made by 31st May 2002 by email to admissions at gatsby.ucl.ac.uk Further information about Wolfram Schultz's lab can be found at http://www.anat.cam.ac.uk/pages/staff/academic/schultz/index.html and about the Gatsby Unit at http://www.gatsby.ucl.ac.uk From jose at psychology.rutgers.edu Mon Apr 22 07:04:39 2002 From: jose at psychology.rutgers.edu (Stephen Hanson) Date: Mon, 22 Apr 2002 07:04:39 -0400 Subject: Postdoctural Position--Please REPLY in subject line RUMBA POSTDOC Message-ID: <3CC3EE47.10202@psychology.rutgers.edu> COGNITIVE/COMPUTATIONAL NEUROSCIENCE POSTDOCTORAL POSITION at RUTGERS UNIVERSITY, Newark Campus. The Rutgers University Mind/Brain Analysis (RUMBA) Project anticipates making one postdoctoral appointment, which is to begin in the SUMMER (June/July) of 2002. This positions are for a minimum of 2 years, with the possibility of continuation for 1 more year and will be in the areas of specialization of cognitive neuroscience with emphasis on the development of new paradigms and methods in neuroimaging, mathematical modeling, signal processing or data analysis in functional brain imaging. Particular interest is in methods and algorithms for fusion of EEG/fMRI. Applications are welcomed begining immediately and review will continue until the position is filled. Rutgers University is an equal opportunity/affirmative action employer. Qualified women and minority candidates are especially encouraged to apply. Send CV and three letters of recommendation and 1 reprint to Professor S.J. Hanson, Department of Psychology, Rutgers University, Newark, NJ 07102. Email enquiry can be made to jose at psychology.rutgers.edu please put "RUMBA POSTDOC" in your subject field also see http://www.rumba.rutgers.edu. From frank.pasemann at ais.fraunhofer.de Tue Apr 23 12:05:49 2002 From: frank.pasemann at ais.fraunhofer.de (Frank Paseman) Date: Tue, 23 Apr 2002 18:05:49 +0200 Subject: Research Associate - Evolutionary Robotics Message-ID: <3CC5865D.20980EAE@ais.fraunhofer.de> I would be grateful if you could distribute the following announcement: The Fraunhofer Institute for Autonomous Intelligent Systems, Sankt Augustin, Germany, invites applications for a Research Associate position in the area of evolutionary robotics. The project will focus on the evolution of modular neurocontrollers for autonomous systems combined with the implementation of behavior relevant learning rules for general recurrent neural networks. Experiments will involve different robot platforms like Kheperas, quadruped and hexapod robots, holonomic robots, etc. The more theoretically oriented activities will be concerned with the relationship between robot behavior, network structure, and network dynamics, theory of evolutionary processes and optimization of evolutionary search strategies, behavior oriented on-line learning rules. The ideal candidate should have strong background in one or more of the following areas: dynamical systems theory, complex systems theory, control theory, recurrent neural networks, behavior based robotics, embodied cognitive science. He should have strong programming skills in C/C++ under a Linux environment. Experience with programming simulators for robots and physical environments will be an advantage. Applicants should have a university degree (Masters or german diploma) in computer science, physics or mathematics. Funding is available for 3 years. The salary will be on the BAT-IIa scale. The post is tenable immediately, but start time is negotiable. For further information and informal enquiries please contact Prof. F. Pasemann (frank.pasemann at ais.fraunhofer.de/http://www.ais.fraunhofer.de/INDY). ************************************************************** Prof. Dr. Frank Pasemann Tel: x49-2241-142373 Fraunhofer-Institut Fax: x49-2241-142342 Autonome Intelligente Systeme Schloss Birlinghoven frank.pasemann at ais.fraunhofer.de D-53754 Sankt Augustin http://www.ais.fraunhofer.de/INDY From becker at meitner.psychology.mcmaster.ca Wed Apr 24 09:28:31 2002 From: becker at meitner.psychology.mcmaster.ca (S. Becker) Date: Wed, 24 Apr 2002 09:28:31 -0400 (EDT) Subject: predoc/doctoral/postdoc positions in auditory neuroscience Message-ID: RESEARCH POSITIONS AND TRAINING OPPORTIUNITES IN AUDITORY MODELLING, PSYCHOPHYSICAL TESTING, AND PLASTICITY IN THE AUDITORY SYSTEM AT MCMASTER UNIVERSITY We are seeking candidates for the following three positions in human auditory neuroscience at McMaster University. Students interested in PhD training in the these fields are also invited to contact us. (1) COMPUTATIONAL MODELS OF HEARING LOSS Postdoctoral and doctoral candidates are sought to develop computational models of sub-cortical and cortical reorganization after noise-induced hearing loss. Both positions will be based at McMaster, in the departments of Psychology and Electrical and Computer Engineering. Excellent programming skills are required. Experience with neural network modelling and signal processing and knowledge of the auditory system and psychoacoustics are highly desirable. The postdoctoral position requires a PhD in Cognitive Science, Computer Science, Electrical Engineering or a related discipline. This research is part of a multi-disciplinary project to study mechanisms and treatment of tinnitus, funded by a $1.1M CIHR grant (NET Programme) to L. Roberts and S. Becker (Psychology, McMaster University), I. Bruce (Electrical and Computer Engnineering, McMaster University), J. Eggermont (Faculty of Medicine, University of Calgary), C. Pantev (Rotman Research Institute, Toronto) and L. Ward (Psychology, University of British Columbia). Interested candidates should send a letter of intention and CV via email with subject line "tinnitus postdoc/doctoral positions" to both Dr. S. Becker (becker at mcmaster.ca) and Dr. I. Bruce (ibruce at ieee.org). (2) PSYCHOPHYSICAL ASSESSMENT OF AUDITORY FUNCTION A Research Associate is sought to conduct psychophysical tests of speech intelligibility in normal and hearing-impaired participants, in order to evaluate hearing compensation algorithms. Experience in psychophysical testing, computer programming, and signal processing are highly desirable. This work is part of a collaborative project to develop intelligent signal processing and neural network algorithms for hearing aids, funded by a $0.5M Collaborative Research Opportunities grant from NSERC, Canada to a multi-disciplinary team of researchers at McMaster University: S. Haykin (Electrical and Computer Engineering), L. Trainor, S. Becker. R. Racine and J. Platt (Psychology). Interested candidates should send a letter of intention and CV via email with subject line "psychoacoustics RA position" to Dr. L. Trainor, ljt at mcmail.cis.mcmaster.ca (3) NEURAL PLASTICITY IN THE AUDITORY SYSTEM We are seeking a scientific colleague to work with us on evaluating models of residual inhibition in tinnitus and on EEG studies of neural plasticity in the human auditory system. The candidate should be a self-directed individual who can work effectively in a collaborative research environment. Strong computing skills and a working knowledge of the auditory system and signal processing are important assets. This position is available at the postdoctoral or similar level, although senior PhD students with suitable background may be considered. The research is part of a multi-site CIHR NET Progamme on understanding, treating, and preventing tinnitus, which is based at McMaster University and three other institutions (see position #1 above). Interested candidates should send their curriculum vitas and statement of interest by email to Dr. L.E. Roberts (roberts at mcmaster.ca). PLease put the phrase "Auditory Scientist Position" in the subject line of your email message. For further information on the research interests of the team see: Tinnitus project: www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.psychology.mcmaster.ca/hnplab www.ucalgary.ca/UofC/faculties/medicine/PHBI/faculty/eggermont.html www.rotman-baycrest.on.ca/content/people/profiles/pantev.html http://neuron2.psych.ubc.ca/~lward/people/index.html Intelligent hearing aids project: www.science.mcmaster.ca/Psychology/http://www.psychology.mcmaster.ca/ljt/ www.crl.mcmaster.ca/People/Faculty/Haykin/haykin.html www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.science.mcmaster.ca/Psychology/racine-web-page/racine.html www.science.mcmaster.ca/Psychology/platt/jp.html From bogus@does.not.exist.com Wed Apr 24 04:20:44 2002 From: bogus@does.not.exist.com () Date: Wed, 24 Apr 2002 01:20:44 -0700 Subject: AI Statistics Conference, Key West, January 2003: Preliminary call for papers Message-ID: <3C0C2AFA8DA9F5448E401726A56637F6AE56E7@RED-MSG-10.redmond.corp.microsoft.com> PRELIMINARY CALL FOR PAPERS Ninth International Conference on Artificial Intelligence and Statistics January 3-6, 2003, Hyatt Hotel, Key West, Florida http://research.microsoft.com/conferences/AIStats2003/ This is the ninth in a series of workshops which have brought together researchers in Artificial Intelligence and in Statistics to discuss problems of mutual interest. The exchange has broadened research in both fields and has strongly encouraged interdisciplinary work. Papers on all aspects of the interface between AI & Statistics are encouraged. Deadlines: Submission of full length 4-page papers for review: 2 August 2002 Submission of final camera-ready papers: 1 November 2002 Invited Speakers: Andrew Blake (Microsoft Research, Cambridge) Bill Freeman (MIT) Zoubin Ghahramani (Gatsby Computational Neuroscience Unit) David Haussler (UCSC) Geoffrey Hinton (University of Toronto) Tommi Jaakkola (MIT) Larry Saul (University of Pennsylvania) Key West provides a superb location for this workshop, and the weather at Key West in January is expected to be very pleasant. The workshop timetable will focus on morning and early evening sessions, allowing ample free time in the afternoons for scientific discussions or to take advantage of local attractions such as scuba diving, snorkelling, wave runners, parasailing, fishing, sailing and golf. From rich at richsutton.com Thu Apr 25 18:53:22 2002 From: rich at richsutton.com (Rich Sutton) Date: Thu, 25 Apr 2002 18:53:22 -0400 Subject: RL [Reinforcement Learning] FAQ Message-ID: Dear Connectionists, I have made a basic RL FAQ available at http://richsutton.com/RL-FAQ.html. Additions/corrections/improvements are encouraged. Rich Sutton From cindy at bu.edu Thu Apr 25 15:34:38 2002 From: cindy at bu.edu (Cynthia Bradford) Date: Thu, 25 Apr 2002 15:34:38 -0400 Subject: 6th ICCNS: Call for Registration Message-ID: <200204251934.g3PJYcD04549@cns-pc75.bu.edu> Apologies if you receive this more than once. ***** CALL FOR REGISTRATION ***** ***** AND ***** ***** FINAL INVITED PROGRAM ***** SIXTH INTERNATIONAL CONFERENCE ON COGNITIVE AND NEURAL SYSTEMS Tutorials: May 29, 2002 Meeting: May 30 - June 1, 2002 Boston University http://www.cns.bu.edu/meetings/ This interdisciplinary conference focuses on two fundamental questions: How Does the Brain Control Behavior? How Can Technology Emulate Biological Intelligence? A single oral or poster session enables all presented work to be highly visible. Contributed talks will be presented on each of the three conference days. Three-hour poster sessions with no conflicting events will be held on two of the conference days. All posters will be up all day, and can also be viewed during breaks in the talk schedule. CONFIRMED INVITED SPEAKERS TUTORIAL SPEAKERS: Wednesday, May 29, 2002 Mark Gluck (Rutgers University) Neural networks in neurology and clinical neuropsychology: Alzheimer's disease, amnesia, and Parkinson's disease Gail A. Carpenter (Boston University) Adaptive resonance theory Ferdinando Mussa-Ivaldi (Northwestern University Medical School) Learning and adaptive control of arm movements Frank Guenther (Boston University) A model of the neural bases of speech motor control INVITED SPEAKERS Thursday, May 30, 2002 CELL AND CIRCUIT DYNAMICS: Daniel Johnston (Baylor College of Medicine) Information processing and storage by neuronal dendrites Bard Ermentrout (University of Pittsburgh) The role of oscillations in odor learning in the Limax John Rinzel (New York University) Cellular dynamics involved in sound localization VISION AND IMAGE PROCESSING: Rudiger von der Heydt (Johns Hopkins University School of Medicine) Global structure in local feature maps David J. Field (Cornell University) Visual systems and the statistics of natural scenes: How far can we go? Philip J. Kellman (UCLA) From jebara at cs.columbia.edu Fri Apr 26 21:55:55 2002 From: jebara at cs.columbia.edu (Tony Jebara) Date: Fri, 26 Apr 2002 18:55:55 -0700 Subject: ICDL'02 - Development and Learning Message-ID: Dear Connectionists, You may be interested in attending the following conference. Please note that early registration rates are available until May 5th, 2002. Details and the program are available at: http://www.egr.msu.edu/icdl02/ Apologies if you receive multiple copies of this message. Thanks... -Tony Jebara ------------------------------------------------------------------------- The 2nd International Conference on Development and Learning (ICDL'02) June 12 - 15, 2002 Massachusetts Institute of Technology Cambridge, Massachusetts USA http://www.egr.msu.edu/icdl02/ Sponsored by: American Association for Artificial Intelligence Cognitive Science Society IEEE Computer Society IEEE Neural Networks Society IEEE Robotics and Automation Society Recent advances in artificial intelligence, cognitive science, neuroscience and robotics have stimulated the birth and growth of a new research field, known as computational autonomous mental development. Although human mental development is a well known subject of study, e.g., in developmental psychology, computational studies of mental development for either machines or humans had not received sufficient attention in the past. Mental development is a process during which a brain-like natural or artificial embodied system, under the control of its intrinsic species-specific developmental program residing in the genes or artificially designed, develops mental capabilities through its autonomous real-time interactions with its environments (including its own internal environment and components) using its own sensors and effectors. The scope of mental development includes cognitive, behavioral, emotional and all other mental capabilities that are exhibited by humans, higher animals and artificial systems. Investigations of the computational mechanisms of mental development are expected to improve our systematic understanding of the working of the wide variety of cognitive and behavioral capabilities in humans and to enable autonomous development of these highly complex capabilities by robots and other artificial systems. Organizers: James L. McClelland Alex P. Pentland Jeff Elman Mriganka Sur Juyang Weng Sridhar Mahadevan Tony Jebara Program Overview: June 12, 2002: Cortical Development and Learning during Vision, Recognition, and Action (Stephen Grossberg) Autonomous Mental Development for Robots (Juyang Weng) June 13, 2002: Cortical Plasticity Contributing to Variations in Human Performance Ability (Michael Merzenich) A Developing Sensory Mapping for Robots (N. Zhang, J. Weng, and Z. Zhang) Combining Embodied Models and Empirical Research for Understanding the Development of Shared Attention (I. Fasel, G. Deak, J. Triesch, and J. Movellan) Learning to Recognize Human Action Sequences (C. Yu, and D. Ballard) The Development of Gaze Following as a Bayesian Systems Identification Problem (J. Movellan, and J.S. Watson) Prediction-error Driven Learning: The Engine of Change in Cognitive Development (James McClelland) Combining Configural and TD Learning on a Robot (D.S. Touretzky, N.D. Daw, and E.J. Tira-Thompson) Action Chaining by a Developmental Robot with a Value System (Y. Zhang, and J. Weng) A Model of an Expectancy-driven and Analogy-making Actor (A. Kulakov, G. Stojanov, and D. Davcec) How do Features of Sensory Representations Develop? (Jon Kass) Dav: A Humanoid Robot Platform for Autonomous Mental Development (J.D. Han, S.W. Zeng, K.Y. Tham, M. Badgero, and J. Weng) A Novel Optimal Discriminant Principle in High Dimensional Spaces (Y. Guo, and L. Wu) Cortical Software Re-Use: A Computational Principle for Cognitive Robotics (R. Reilly, and I. Marian) Contentful Mental States for Robot Baby (P.R. Cohen, J.T. Oates, and C.R. Beal) An Incremental Representation of Conceptual Symbols Using RCE Neural Network (M. Xie, and M.L.Yuan) A Developmental Principle for Robotic Hand-Eye Coordination Skill (M. Xie) An Evolutional Network Architecture for Developmental Knowledge Bases (A. Saad, and A.R.M. Zaghloul) Towards a Theory Grounded Theory of Language (C.G. Prince, E.J. Mislivec, O.V. Kosolapaov, and T.R. Lykken) June 14,2002: One Thing Follows Another: Initial State, Task, and Developmental Change in Human Infants (Esther Thelen) A Theory for Mentally Developing Robots (J. Weng) Many-Layered Learning (P.E. Utgoff, and D.J. Stracuzzi) Beyond the Turing Test: Performance Metrics for Evaluating a Computer Simulation of the Human Mind (N. Alvarado, S. S. Adams, S. Burbeck, and C. Latta ) Learning in Content-based Image Retrieval (Thomas S. Huang) Developmental Learning of Memory-based Perceptual Models (Y. Ivanov, and B. Blumberg) Learning to Detect Multi-View Faces in Real-Time (S.Z. Li, L. Zhu, Z.Q,. Zhang, and H.T. Zhang) Local Non-Negative Matrix Factorization as a Visual Representation (T. Feng, S.Z. Li, H.Y. Shum, and H.J. Zhang) Humanoid Robot Models of Child Development (Rodney Brooks) Statistical Imitative Learning from Perceptual Data (T. Jebara, and A. Pentland) Learning Prospective Pick and Place Behavior (D. Wheeler, A.H. Fagg, and R.A. Grupen) Learning Movement Sequences from Demonstration (A. Ramesh, and M.J. Mataric) June 15, 2002: Rewiring Cortex: Rules of Cortical Network Development (Mriganka Sur) Interactions Between Development and Learning During the Acquisition of Binocular Disparity Sensitivities (M. Dominguez, and R.A. Jacobs) Dynamic Growth Modeling of Human Cognitive Microdevelopment (Z. Yan, and K. Fischer) Marginal Self-organization: A Model of the Role of Executive Processes in Learning (R. Viviani) Learning Your Life: Wearables and Familiars (Alex Pentland) On the Development of Visual Object Memory: The Stay/Go Decision Problem (C.T. Morrison, P. Cohen, and P. Sebastiani) A Computer-based Tutoring System for Visual-Spatial Skills: Dynamically Adapting to the User?s Developmental Range (M.W. Connell, and D.A. Stevens) Extending the BDI Model to Accelerate the Mental Development of Autistic Patients (B. Galitsky) Development as a Source of Complexity (Jeff Elman) Dopamine, Reward Conditioning and Robot Behavior (O. Sporns, and W.H. Alexander) Dopamine and Inference About Timing (N. Daw, A.C. Courville, and D.S. Touretzky) A Developmental Approach Accelerates Learning of Joint Attention (Y. Nagai, M. Asada, and K. Hosoda) From laubach at kafka.med.yale.edu Sun Apr 28 21:15:02 2002 From: laubach at kafka.med.yale.edu (Mark Laubach) Date: Sun, 28 Apr 2002 21:15:02 -0400 Subject: PROGRAMMING THE "BRAIN-MACHINE INTERFACE" Message-ID: PROGRAMMING THE "BRAIN-MACHINE INTERFACE" A position is available at the John B. Pierce Laboratory, affiliated with the Yale School of Medicine, for someone with strong computer programming skills (C/C++, Python, Matlab, S+/R). This person will work as part of a team of neuroscientists, engineers, mathematicians, and computer scientists working on the "brain-machine interface" (sponsored by a grant from DARPA). This position requires knowledge of UNIX-style operating systems, modern methods for data analysis and machine learning, and distributed and/or real-time computing, is available starting July 1, 2002, and is paid at NIH-standard levels. If interested, please send a vita, a summary of skills and experience, and the names of three references to: Mark Laubach, Ph.D. The John B. Pierce Laboratory Yale School of Medicine 290 Congress Ave New Haven CT 06519 E-mail: laubach at kafka.med.yale.edu http://www.jbpierce.org/staff/laubach.html ------ NOTE: This is a follow-up to a recent posting related to this project. Three of the other positions advertised in our original posting have now been filled. From mpessk at guppy.mpe.nus.edu.sg Mon Apr 29 01:00:42 2002 From: mpessk at guppy.mpe.nus.edu.sg (S. Sathiya Keerthi) Date: Mon, 29 Apr 2002 13:00:42 +0800 (SGT) Subject: Extensions of the SMO algorithm Message-ID: Dear Connectionists: We have recently completed two papers on the extensions of the SMO algorithm to Least Squares SVM formulations and Kernel Logistic Regression. Gzipped postscript files containing these papers can be downloaded from: http://guppy.mpe.nus.edu.sg/~mpessk/svm.shtml The titles and abstracts of these papers are given below. S. Keerthi -------------------------------------------------------------- SMO Algorithm for Least Squares SVM Formulations S.S. Keerthi and S.K. Shevade This paper extends the well-known SMO algorithm of Support Vector Machines (SVMs) to Least Squares SVM formulations which include LS-SVM classification, Kernel Ridge Regression and a particular form of regularized Kernel Fisher Discriminant. The algorithm is shown to be asymptotically convergent. It is also extremely easy to implement. Computational experiments show that the algorithm is fast and scales efficiently (quadratically) as a function of the number of examples. -------------------------------------------------------------- A Fast Dual Algorithm for Kernel Logistic Regression S.S. Keerthi, K. Duan, S.K. Shevade and A.N. Poo (Accepted for presentation at ICML 2002) This paper gives a new iterative algorithm for kernel logistic regression. It is based on the solution of the dual problem using ideas similar to those of the SMO algorithm for Support Vector Machines. Asymptotic convergence of the algorithm is proved. Preliminary computational experiments show that the algorithm is robust and fast. The algorithmic ideas can also be used to give a fast dual algorithm for solving the optimization problem arising in the inner loop of Gaussian Process classifiers. -------------------------------------------------------------- From amy at cs.umass.edu Mon Apr 29 17:39:10 2002 From: amy at cs.umass.edu (Amy McGovern) Date: Mon, 29 Apr 2002 17:39:10 -0400 Subject: Ph.D. thesis available Message-ID: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Dear Connectionists, I am pleased to announce the availability of my Ph.D. thesis: Autonomous Discovery of Temporal Abstractions from Interaction with an Environment Amy McGovern University of Massachusetts Amherst The thesis is available from either of the following links: http://www-anw.cs.umass.edu/~amy/pubs/mcgovern_thesis.pdf http://www-anw.cs.umass.edu/~amy/pubs/mcgovern_thesis.ps --------------------------------- Abstract The ability to create and to use abstractions in complex environments, that is, to systematically ignore irrelevant details, is a key reason that humans are effective problem solvers. Although the utility of abstraction is commonly accepted, there has been relatively little research on autonomously discovering or creating useful abstractions. A system that can create new abstractions autonomously can learn and plan in situations that its original designer was not able to anticipate. This dissertation introduces two related methods that allow an agent to autonomously discover and create temporal abstractions from its accumulated experience with its environment. A temporal abstraction is an encapsulation of a complex set of actions into a single higher-level action that allows an agent to learn and plan while ignoring details that appear at finer levels of temporal resolution. The main idea of both methods is to search for patterns that occur frequently within an agent's accumulated successful experience and that do not occur in unsuccessful experiences. These patterns are used to create the new temporal abstractions. The two types of temporal abstractions that our methods create are 1) subgoals and closed-loop policies for achieving them, and 2) open-loop policies, or action sequences, that are useful ``macros.'' We demonstrate the utility of both types of temporal abstractions in several simulated tasks, including two simulated mobile robot tasks. We use these tasks to demonstrate that the autonomously created temporal abstractions can both facilitate the learning of an agent within a task and can enable effective knowledge transfer to related tasks. As a larger task, we focus on the difficult problem of scheduling the assembly instructions for computers with multiple pipelines in such a manner that the reordered instructions will execute as quickly as possible. We demonstrate that the autonomously discovered action sequences can significantly improve performance of the scheduler and can enable effective knowledge transfer across similar processors. Both methods can extract the temporal abstractions from collections of behavioral trajectories generated by different processes. In particular, we demonstrate that the methods can be effective when applied to collections generated by reinforcement learning agents, heuristic searchers, and human tele-operators. From juergen at idsia.ch Tue Apr 30 07:34:58 2002 From: juergen at idsia.ch (Juergen Schmidhuber) Date: Tue, 30 Apr 2002 13:34:58 +0200 Subject: Ph.D. thesis on temporal abstraction References: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Message-ID: <3CCE8162.11399073@idsia.ch> Re: Amy McGovern's announcement of "Autonomous Discovery of Temporal Abstractions from Interaction with an Environment" Let us not forget Mark Ring's early work on temporal abstraction for reinforcement learning, e.g: Mark. B. Ring: Incremental Development of Complex Behaviors through Automatic Construction of Sensory-Motor Hierarchies. Machine Learning: Proceedings of the Eighth International Workshop, In L. Birnbaum and G. Collins, eds., 343-347, Morgan Kaufmann, 1991. Additional algorithms in his NIPS 5 paper (1993) and his 1994 thesis. Maybe also check out http://www.idsia.ch/~juergen/subgoals.html Juergen Schmidhuber http://www.idsia.ch/~juergen From h.wang at ulster.ac.uk Tue Apr 30 07:55:01 2002 From: h.wang at ulster.ac.uk (Hui Wang) Date: Tue, 30 Apr 2002 12:55:01 +0100 Subject: Special Issue of Decision Support Systems on Data Mining for Financial Decision Making Message-ID: <072701c1f03d$dbe8e870$a2943dc1@tiger> Dear Connectionists, You may be interested in the following special issue of Decision Support Systems. Details are available at http://www.weigend.com/dss and http://www.elsevier.com/inca/homepage/sae/orms/dss/call1.htm Apologies if you receive multiple copies of this message. Thanks... Hui Wang Andreas Weigend ------------------------------------------------------------------------------- The Journal of Decision Support Systems Special Issue on Data Mining for Financial Decision Making GUEST EDITORS Hui Wang, University of Ulster Andreas S. Weigend, Weigend Associates LLC CALL FOR PAPERS As information intensive organizations transform themselves from passive collectors to active explorers and exploiters of data, they face a serious challenge: How can they benefit from increased access to information to better understand their markets, customers, suppliers, operations and internal business processes? Responding to this challenge, the field of data mining has emerged. It focuses on the process of discovering valid, comprehensible, and potentially useful knowledge from large data sets with the goal to apply this knowledge to decision making. Data mining integrates concepts from modern statistics, intelligent information systems, machine learning, pattern recognition, decision theory, data engineering and database management, and provides powerful tools that can reveal complex and hidden relationships in large amounts of data. The approaches include neural networks, genetic programming, and tree-based methods. Data mining already has a major impact on business and finance. Financial markets generate large volumes of data. Analysing these data to reveal valuable information and making use of the information in decision making present great opportunities but grand challenges for data mining. The rewards for finding valuable patterns are potentially enormous, but so are the difficulties. There is evidence that short-term trends do exist and some general patterns do occur frequently. Important problems are: how to find the trends at their early stages and how to time the beginning and ending of trends, how to take into account in decision making the found trends, the general patterns, and domain knowledge that describes the intricately inter-related world of global financial markets. The focus of this special issue is on the use of data mining techniques for decision making in financial markets. Topics of interest include: * Financial data selection and pre-processing for data mining * Solutions to new problems in financial decision making * New solutions for classical problems in financial decision making * Data and solutions visualisation for financial decision making * Successful case studies. Areas include: * Risk management including credit risk and market risk * Asset allocation, dynamic trading and hedging * Execution and liquidity models * Behavioural finance, and other emerging areas. Both original contributions and thoughtful survey papers are welcome. SUBMISSION INSTRUCTIONS Electronic submissions are strongly encouraged. Postscript or PDF copies of manuscripts may be emailed to h.wang at ulst.ac.uk. SUBMISSION DEADLINE: September 9, 2002 Details about the submission process and scope of the special issue are available at http://www.weigend.com/dss and http://www.elsevier.com/inca/homepage/sae/orms/dss/call1.htm Hui Wang School of Information and Software Engineering University of Ulster at Jordanstown Northern Ireland, BT37 0QB United Kingdom Tel: +44 28 90368981 Fax: +44 28 90366068 Email: h.wang at ulst.ac.uk Andreas S. Weigend Weigend Associates LLC P.O.Box 20207 Stanford, CA 94309 U.S.A. Tel: +1 917 697-3800 Fax: +1 815 327-5462 Email: dss at weigend.com From juergen at idsia.ch Tue Apr 30 07:34:58 2002 From: juergen at idsia.ch (Juergen Schmidhuber) Date: Tue, 30 Apr 2002 13:34:58 +0200 Subject: Ph.D. thesis on temporal abstraction References: <200204292139.g3TLdBb22772@ananke.cs.umass.edu> Message-ID: <3CCE8162.11399073@idsia.ch> Re: Amy McGovern's announcement of "Autonomous Discovery of Temporal Abstractions from Interaction with an Environment" Let us not forget Mark Ring's early work on temporal abstraction for reinforcement learning, e.g: Mark. B. Ring: Incremental Development of Complex Behaviors through Automatic Construction of Sensory-Motor Hierarchies. Machine Learning: Proceedings of the Eighth International Workshop, In L. Birnbaum and G. Collins, eds., 343-347, Morgan Kaufmann, 1991. Additional algorithms in his NIPS 5 paper (1993) and his 1994 thesis. Maybe also check out http://www.idsia.ch/~juergen/subgoals.html Juergen Schmidhuber http://www.idsia.ch/~juergen From mpessk at guppy.mpe.nus.edu.sg Mon Apr 29 01:00:42 2002 From: mpessk at guppy.mpe.nus.edu.sg (S. Sathiya Keerthi) Date: Mon, 29 Apr 2002 13:00:42 +0800 (SGT) Subject: Extensions of the SMO algorithm Message-ID: Dear Connectionists: We have recently completed two papers on the extensions of the SMO algorithm to Least Squares SVM formulations and Kernel Logistic Regression. Gzipped postscript files containing these papers can be downloaded from: http://guppy.mpe.nus.edu.sg/~mpessk/svm.shtml The titles and abstracts of these papers are given below. S. Keerthi -------------------------------------------------------------- SMO Algorithm for Least Squares SVM Formulations S.S. Keerthi and S.K. Shevade This paper extends the well-known SMO algorithm of Support Vector Machines (SVMs) to Least Squares SVM formulations which include LS-SVM classification, Kernel Ridge Regression and a particular form of regularized Kernel Fisher Discriminant. The algorithm is shown to be asymptotically convergent. It is also extremely easy to implement. Computational experiments show that the algorithm is fast and scales efficiently (quadratically) as a function of the number of examples. -------------------------------------------------------------- A Fast Dual Algorithm for Kernel Logistic Regression S.S. Keerthi, K. Duan, S.K. Shevade and A.N. Poo (Accepted for presentation at ICML 2002) This paper gives a new iterative algorithm for kernel logistic regression. It is based on the solution of the dual problem using ideas similar to those of the SMO algorithm for Support Vector Machines. Asymptotic convergence of the algorithm is proved. Preliminary computational experiments show that the algorithm is robust and fast. The algorithmic ideas can also be used to give a fast dual algorithm for solving the optimization problem arising in the inner loop of Gaussian Process classifiers. --------------------------------------------------------------