From zhaoping at gatsby.ucl.ac.uk Mon Feb 1 06:49:40 1999 From: zhaoping at gatsby.ucl.ac.uk (Dr Zhaoping Li) Date: Mon, 1 Feb 1999 11:49:40 GMT Subject: Ph.D. studentships in Computational Neuroscience at Gatsby Unit Message-ID: <199902011149.LAA15879@vision.gatsby.ucl.ac.uk> Applications are invited for research studentships in the field of computational neuroscience tenable from September 1999. In particular, applications are welcome from candidates with an interest to pursue research in the area of high and low level vision and/or motor control. Applicants should have a strong analytical background, a good honours degree in a relevant subject, and a keen interest in neuroscience. Competitive studentships are available. To apply, email a statement of your research interest and a text-only version of your CV to Janice Hankes at admissions at gatsby.ucl.ac.uk by March 1 1999. Further details about the Gatsby Unit can be found at http://www.gatsby.ucl.ac.uk. From smagt at dlr.de Tue Feb 2 06:51:38 1999 From: smagt at dlr.de (Patrick van der Smagt) Date: Tue, 2 Feb 1999 12:51:38 +0100 (MET) Subject: Announcement: IEE Workshop on cerebellar control (London, Feb. 15) Message-ID: <199902021151.MAA07667@ilz.robotic> WORKSHOP ANNOUNCEMENT ===================== Self-learning robots III Brainstyle robotics: The cerebellum beyond function approximation http://www.robotic.dlr.de/Smagt/workshop/iee99.html A Workshop organised by Professional Group A9 (Neural computing) Monday 15 February 1999, Savoy Place, London, UK Patrick van der Smagt (Institute of Robotics and System Dynamics, DLR), workshop co-organiser Noel Sharkey (University of Sheffield), workshop co-organiser Throughout the years there has been extensive research in the structure and working of the cerebellum. The cerebellum is a part of the brain of vertebrates that is well understood in various aspects: The cerebellum is used for the spatial motor coordination of fast movements. High precision of the movement is an important issue. The structure of the cerebellum is well-understood: the cells as well as their connections are known. Also the connections to the other parts of the brain and spine are understood. The learning process is well-understood since 1992. The 'traditional' neural approach for robot control typically consists of some form of function approximation. We gather a large amount of learning samples, and try to represent those as well as possible with some (universal or specialized) approximator. The major differences lie there where the choice of approximator is concerned. Does the cerebellum do something different? Many special issues and workshops focusing on brain models for robot control reflect the increased attention of the development of cerebellar models. But what can we expect of such models in the future? Is the cerebellum is 'just' a function approximator, and can we mimic its functionality using other models? In this workshop we will focus on 'working' models of the cerebellum; how far have we come? Can real applications be expected in the near future? Can cerebellar models do better than others? Are brainstyle robotics something to expect in the near future? Programme 09.00 Registration and Coffee 09.30 Welcome 09.40 Overview of cerebellar control Patrick van der Smagt (Institute of Robotics and System Dynamics, DLR) 10.20 The cerebellum and visually controlled movements Chris Miall (University Laboratory of Physiology, Oxford) 11.00 The cerebellum as a neuronal machine David Willshaw (University of Edinburgh) 11.40 Recurrent disinhibition implements a pointer-map module between cerebellar basket and Purkinje cells Klaus Hepp (Institute of Theoretical Physics, ETH Z"urich) 12.20 Lunch 13.20 Computational Delays and Habits Guido Bugmann (University of Plymouth) 14.00 Learning to Reach Via Corrective Movements Andrew Barto (University of Massachusetts) 15.00 Tea 15.15 Cerebellar vs. stiffness control Pietro Morasso (Universit'a di Genova) 15.55 Panel discussion 16.30 Close To register for the above event, please contact: Events Office, IEE, Savoy Place, London WC2R 0BL, tel: +44 (0)171 240 1871 ext: 2205/6, fax: +44 (0)171 497 3633 or email: events at iee.org.uk -- dr Patrick van der Smagt phone +49 8153 281152, fax -34 DLR/Institute of Robotics and System Dynamics smagt at dlr.de P.O.Box 1116, 82230 Wessling, Germany http://www.robotic.de/Smagt/ From A.van.Ooyen at nih.knaw.nl Wed Feb 3 09:06:23 1999 From: A.van.Ooyen at nih.knaw.nl (Arjen van Ooyen) Date: Wed, 03 Feb 1999 15:06:23 +0100 Subject: Conference Announcement Message-ID: <36B857DF.3E9F@nih.knaw.nl> Dear Computational Neuroscientist, This mailing is to bring to your attention the International Conference on Theory and Mathematics in Biology and Medicine 1999, that will be held from Tuesday June 29 through Saturday July 3, 1999, in Amsterdam, the Netherlands. At this joint meeting of the European Society for Mathematical and Theoretical Biology (ESMTB), the Society for Mathematical Biology (SMB), and the Netherlands Society for Theoretical Biology (NVTB), a full day is dedicated to Computational Neuroscience. This day will comprise two plenary invited lectures, given by Prof David Willshaw (University of Edinburgh, United Kingdom) and by Prof Idan Segev (Hebrew University, Israel), 12 contributing lectures of 20 minutes duration, as well as a poster session. Please find all information of the conference at http://www.bio.vu.nl/tmbm99/ We hope that the opportunity to emphasize Computational Neuroscience in the broad field of Mathematical and Theoretical Biology as well as the opportunity to become informed about the progress in other interesting topics at this meeting will encourage you to participate and submit an abstract. ******************************************************* The deadline of Abstract Submission is 1st March, 1999. ******************************************************* Kind regards, and we look forward to meeting you this summer in Amsterdam, Jaap van Pelt, Arjen van Ooyen Netherlands Institute for Brain Research, Amsterdam, The Netherlands Organisers Session Computational Neuroscience Brief overview of the Computational Neuroscience Session and the rest of the conference: ---------------------------------------------------------------- Understanding the brain in health and disease is perhaps the greatest challenge for science in the next century. Because the brain is an extremely complex and organised system from the molecular up to the behavioural level, this goal can not be achieved without modelling and computational approaches. The Computational Neuroscience Session aims at emphasizing this role of Mathematical Modeling and Computer Simulation in Neuroscience. Participants are invited to present progress in understanding the structure, function and development the brain. ----------------------------------------------------------------- The full-day major topics of this conference include 1. Non-Linear Population Dynamics 2. Epidemiology of Infectious Diseases 3. Sequencing, Folding and Molecular Structure 4. Evolution, Game Theory and Adaptive Dynami 5. Computational Neuroscience 6. Immunology and Within-Host Dynamics of Pathogens The half-day minor topics of this conference include 1. Individual-Based Population Dynamics 2. Gene Networks 3. Pattern Formation and Morphogenesis 4. Education in Mathematical Biology 5. Parasites and Diseases in Wildlife 6. Metabolic Control Networks 7. Whole Heart Modelling 8. Metapopulation Dynamics 9. Toxicology 10. Particle Based Modelling 11. Networks, Cellular Signalling and Biological Rhythms 12. Classification Methodology 13. Miscellaneous ========================================================== -- Arjen van Ooyen, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. email: A.van.Ooyen at nih.knaw.nl website: http://www.cns.ed.ac.uk/people/arjen.html phone: +31.20.5665483 fax: +31.20.6961006 From john at dcs.rhbnc.ac.uk Wed Feb 3 12:34:49 1999 From: john at dcs.rhbnc.ac.uk (John Shawe-Taylor) Date: Wed, 3 Feb 1999 17:34:49 +0000 (GMT) Subject: Informal pre-EuroCOLT'99 Workshop on Kernel Methods In-Reply-To: <01BE0EFC.3AF90BA0.jkh@dcs.rhbnc.ac.uk> Message-ID: We are hosting a one day informal workshop on Sunday 28th March at Nordkirchen Castle, Germany, on the Sunday before the EuroCOLT'99 conference; particular interest of the organisers is the analysis of Kernel Methods and this will be one of the themes of the workshop. To find out more information and registration procedures please go to the web site: http://svm.first.gmd.de/eurocolt99/workshop.html Organizing Committee John Shawe-Taylor, john at dcs.rhbnc.ac.uk Alex J. Smola, smola at first.gmd.de Bernhard. Schoelkopf, bs at first.gmd.de Robert C. Williamson, Bob.Williamson at anu.edu.au From Zoubin at gatsby.ucl.ac.uk Thu Feb 4 08:17:22 1999 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 4 Feb 1999 13:17:22 +0000 (GMT) Subject: Gatsby Unit: PhD and Postdoc positions Message-ID: <199902041317.NAA18273@cajal.gatsby.ucl.ac.uk> University College London Gatsby Computational Neuroscience Unit Current vacancies: PhD and Post-doctoral positions ------------------------------ Post-doctoral and PhD Research Positions High and low level vision Applications are invited for a post-doctoral research fellowship and a PhD research studentship in the field of high and low level vision tenable from September 1999. The successful candidate/s will be supervised by Dr Zhaoping Li. Applicants should have a strong analytical background, and a keen interest in neuroscience. Competitive salaries and studentships are available. ------------------------------ PhD Research Studentship Computational and experimental motor control Applications are invited for a PhD research studentship in the field of computational and experimental motor control tenable from September 1999. The successful candidate will be supervised by Dr Zoubin Ghahramani. Applicants should have a strong analytical background, and a keen interest in neuroscience. Competitive salaries and studentships are available. ------------------------------ APPLICATION DEADLINE: Monday 1 March 1999 HOW TO APPLY: http://www.gatsby.ucl.ac.uk/vacancies.html EMAIL ENQUIRIES: admissions at gatsby.ucl.ac.uk (Janice Hankes) From horwitz at helix.nih.gov Fri Feb 5 13:40:05 1999 From: horwitz at helix.nih.gov (Barry Horwitz) Date: Fri, 5 Feb 1999 13:40:05 -0500 Subject: Available Postdoctoral Position Message-ID: Postdoctoral Fellowship in Neural Modeling of Human Functional Neuroimaging Data National Institute on Aging and National Institute on Deafness and Other Communication Disorders An entry-level postdoctoral fellowship under the supervision of Dr. Barry Horwitz is available immediately for developing and applying computational neuroscience modeling methods to in vivo human functional neuroimaging data, obtained from positron emission tomography (PET) and functional magnetic resonance imaging (fMRI). The goal is to understand the relation between functional neuroimaging data, with its low spatial and temporal resolution, and the underlying electrophysiological behavior of multiple interconnected neuronal populations. Knowledge of neural modeling techniques and programming experience are required. PET and fMRI facilities are available to implement and test hypotheses derived from the modeling. See Cerebral Cortex 8: 310-320 (1998) for an example of this approach. PhD or MD degree required. The position is in the Language Section, Voice, Speech and Language Branch, NIDCD, NIH, Bethesda, MD USA. For further information, contact: Dr. Barry Horwitz, Bldg. 10, Rm. 6C-414, National Institutes of Health, Bethesda, MD 20892, USA. Tel. 301-594-7755; FAX: 301-402-0074; Email: horwitz at helix.nih.gov. ----------------------------------------------------------------- Barry Horwitz, Ph.D. Tel. 301-594-7755 Bldg. 10, Rm. 6C-414 FAX 301-402-0074 MSC 1588 horwitz at helix.nih.gov Laboratory of Neurosciences National Institute on Aging National Institutes of Health Bethesda, MD 20892 USA From lendaris at sysc.pdx.edu Fri Feb 5 20:12:30 1999 From: lendaris at sysc.pdx.edu (George G. Lendaris) Date: Fri, 05 Feb 1999 17:12:30 -0800 Subject: Director Position: Systems Science Message-ID: <36BB96FD.F83AF184@sysc.pdx.edu> Director, Interdisciplinary Systems Science Ph.D. Program. Looking for a dynamic, creative individual to lead a unique established interdisciplinary Ph.D. program to international prominence, building on resurgent interest in areas such as complexity, chaos, learning and adaptation. The existing program at Portland State University involves a core component and relationships with a dozen academic units with a current enrollment of 145 students. The Systems Science core faculty currently comprises three tenure track and one adjunct faculty. In addition to the general duties associated with directing and administrating the Program, the Director is expected to 1) provide strong interaction with faculty and administration of participating department/schools; 2) cultivate collaborative research and teaching among the faculty in Systems Science, Social Sciences, Natural Sciences, Mathematics, Engineering, and Business; 3) teach as an additional core faculty member; 4) guide the establishment of a new Masters in Systems Science (currently in the approval process); 5) initiate and guide activities to expand funding for research and program development; and 6) direct the student recruiting efforts. Applicants should have an earned doctorate in Systems Science or related discipline and experience in an academic setting including teaching, research publications and grants. Additional information about the Program is available at http://www.sysc.pdx.edu/. Applicants should submit 1) A letter of application; 2) a vita; and 3) three names of reference to Nancy A. Perrin, Chair Search Committee, Portland State University, P.O. Box 751-SYSC, Portland, OR 97207. Review of applications will begin on Feb. 26, 1999, and will continue until position is filled. PSU is an affirmative action/equal opportunity institution. ---------------------------------------------------------------- From bebis at cs.unr.edu Sat Feb 6 18:38:47 1999 From: bebis at cs.unr.edu (George Bebis) Date: Sat, 6 Feb 1999 15:38:47 -0800 (PST) Subject: IEEE International Conference on Information, Intelligence and Systems Message-ID: <199902062338.PAA01323@crete.cs.unr.edu> -------------- next part -------------- First Call for Papers IEEE International Conference on Information, Intelligence and Systems (Symposia) November 1-3, 1999 Washington, DC Sponsored by : IEEE Computer Society In Cooperation with : AAAI Society; SMC Society; NN Society; IAPR Society; ACM Society PAMI TC; Virtual Intelligence TC; TAI Conference; BU-CIS Center, TUCrete and UCrete Conference includes tours to NIST or NASA and NIH or TIGR http://iciis99.cs.unr.edu/ ------------------------------------------------------------------------ The IEEE International Conference on Information, Intelligence, and Systems (ICIIS) offers theoretical and practical media for the constructive interaction among scientists and practitioners from different research fields (computers, mathematics, physics, biology, medicine, chemistry, engineering, etc) having as goals the development of methodologies and tools for the solution of complex problems in neuroscience and biology, automation and robotics, image, speech and natural languages, and their integration. The conference is organized as four simultaneous single-track symposia. General Symposia Chair: N.G.Bourbakis, BU, CIS & UC Email:Bourbaki at Binghamton.edu Phone:(607) 777-2165 ------------------------------------------------------------------------ IEEE Symposium on INTELLIGENCE in NEURAL and BIOLOGICAL SYSTEMS (INBS) SOME TOPICS: Biological Models, Computational Biology, Evolutionary Dynamics, Evolutionary Computation Models, Molecular Biology, DNA Sequence Processing, Genome Processes, DNA Topologies, Genome Mapping and Sequencing, Learning, Perception, Models of Neural Nets, Neuroscientific Models, etc. Program Chairs: J.Gattiker, LANL, and J. Wang, NJIT Phone: (505) 665-0604 & (973) 596-3396 Email: gatt at lanl.gov & jason at village.njit.edu Submit summary to: JR Gattiker TA3, SM43, MSF645, DP13S, XCM Los Alamos National Laboratory Los Alamos, NM 87544 ------------------------------------------------------------------------ IEEE Symposium on INTELLIGENCE in AUTOMATION and ROBOTICS (IAR) SOME TOPICS: Robot Path Planning, Multiple Robot Motion Planning, Autonomous Navigation, Arm/Leg Control, Robot Vision, Visual Tracking, Human-Robot Symbiosis, Telerobotics, Micro and Nano Robotics, Robot Cooperation, Assembly Strategies, Sensors, , Walking/Running Robots, Task Planning, Process Planning, Scheduling, Intelligent Control, etc. Program Chairs: L. Tsoukalas, Purdue U. and A.Tascillo, Ford Phone: (765) 494-0198 & (313) 845-7427 Email: tsoukala at helios.ecn.purdue.edu & atascill at ford.com Submit summary to: L. Tsoukalas 1290 Nuclear Engineering Building Purdue University West Lafayette, IN 47907-1290 ------------------------------------------------------------------------ IEEE Symposium on IMAGE, SPEECH, NATURAL LANGUAGE SYSTEMS (ISNLS) SOME TOPICS: Image (Coding, Compression, Encryption, Segmentation, Enhancement, Restoration, Skeletonization, Analysis, Morphology, Databases, Video, Pattern Recognition, OCR, Image Understanding & Interpretations, Mathematical Methods); Speech (Coding, Compression, Processing, Analysis, Synthesis, Recognition, Understanding); Natural Language (NL) Processing, Computational Linguistics, Document Processing, NL Translation, NL Understanding, Multimedia, etc. Papers on the technical details of the commercial (e.g. , consumer electronics) , scientific, and medical applications of the above are encouraged. Program Chairs: G. Bebis, U. Nevada, and S. Amer, USGS Phone: (702) 784-6463 & (605) 594-6864 Email: bebis at cs.unr.edu & samer at edcmail.cr.usgs.gov Submit summary to: George Bebis Dept of Computer Science University of Nevada Reno, NV 89557 ------------------------------------------------------------------------ IEEE Symposium on INFORMATION and INTELLIGENT AGENTS (IIA) SOME TOPICS: Agent models and architectures, Agent-oriented programming, Communication issues and protocols, Cooperation and coordination, Cooperative information systems, Conflict resolution and negotiation, Distributed search, Intelligent agents, Multiagent planning and learning, Practical applications (Enterprise integration, Feature interaction, Information gathering, Manufacturing, Software agents, Software engineering), Testbeds and development environments, User interface issues. Program Chairs: C. Koutsougeras, Tulane U., and S. Mertoguno, FUJITSU. Phone: (504) 862-3369 & (408) 922-9520 Email: ck at eecs.tulane.edu & jmertohu at fmi.fujitsu.com Submit summary to: Chris Koutsougeras Electrical Engineering and Computer Science Dept. Tulane University New Orleans, LA 70118 ------------------------------------------------------------------------ Information for Authors: Please send four copies of a paper or an extended summary (3-6 pages) describing the methodology and/or results by May 15, 1999 to the appropriate Program Chair's address above. Notification of acceptance (please provide email address or fax number) by June 30, 1999. Camera ready articles by August 15, 1999. Publication of Papers and Awards: This is a fully refereed conference. The proceedings will be published by IEEE Computer Society Press. A collection of best papers will be published in an archival Journal and an Award will be given to each best paper per Symposium. For further details or more up-to-date information please visit the conference WWW page at http://iciis99.cs.unr.edu/ Special Sessions: Proposals are invited for special sessions on any topic relevant to the symposium. If you are interested in organizing an special session please follow the procedure outlined below: 1. Submit a proposal (topic area) to one the appropriate program chair. 2. Upon approval, recruit authors for the session (2 hours, 4/5 papers) and review papers for appropriate topic/contents. 3. Session chair submits abstract *package* of author/title/extended abstracts to the appopriate program chair by May 30. 4. Editorial comments, and official acceptance of the session will be made by the program chair(s) by June 30. From barba at cvs.rochester.edu Mon Feb 8 10:04:41 1999 From: barba at cvs.rochester.edu (Barbara Arnold) Date: Mon, 8 Feb 1999 10:04:41 -0500 Subject: Fellowship program Message-ID: Summer Undergraduate Fellowship Program. The University of Rochester's Center for Visual Science is pleased to announce the 1999 Summer Undergraduate Research Program. Undergraduates in their junior year of baccalaureate studies who have an interest in neurobiological, optical, psychophysical or computational aspects of vision are invited to join a member of our diverse faculty for a summer of supervised laboratory training and participation in vision seminars and journal clubs. Our fellowships will cover travel expenses and provide a stipend of $1000/month for up to 4 months of continuous training that will culminate in participation in the Undergraduate Summer Workshop, Aug 6-9. Rochester, located near Lake Ontario in western New York, is an internationally recognized center for visual and imaging science and provides a scenic and natural setting for a variety of outdoor activities. Applicants will be selected based on academic achievement and a demonstration of interest in pursuing studies in visual science. For application and/or more information, please contact: Barbara Arnold University of Rochester Center for Visual Science RC Box 270270 Rochester, NY 14627-0270 716-275-8659 barba at cvs.rochester.edu.(www.cvs.rochester.edu) Acceptance to Fellowship Program considered upon receipt of application. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Barbara N. Arnold Administrator email: barba at cvs.rochester.edu Center for Visual Science phone: 716 275 8659 University of Rochester fax: 716 271 3043 Meliora Hall 274 Rochester NY 14627-0270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From giles at research.nj.nec.com Mon Feb 8 10:01:01 1999 From: giles at research.nj.nec.com (Lee Giles) Date: Mon, 8 Feb 1999 10:01:01 -0500 (EST) Subject: New paper: neural network classification and prior class probabilities Message-ID: <199902081502.KAA00263@alta> A non-text attachment was scrubbed... Name: not available Type: text Size: 2990 bytes Desc: not available Url : https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/00000000/8b0ffe11/attachment.ksh From avrama at gmu.edu Mon Feb 8 18:20:33 1999 From: avrama at gmu.edu (avrama) Date: Mon, 8 Feb 1999 18:20:33 -0500 (EST) Subject: PhD Fellowship Message-ID: George Mason University Institute for Computational Sciences and Informatics Ph.D. Research Position ------------------------------ Applications are invited to the Ph.D. program in Bioinformatics, Computational Neuroscience concentration. Fellowship money is available to work with Dr. Blackwell, a CSI faculty member and affiliated with the Krasnow Institute. The area of research is the biophysical basis of long term memory. Applicants are expected to have experience in C programming and UNIX operating systems; completed course work on Ordinary differential equations and linear algebra; and an interest in neuroscience. Applications may be downloaded from http://csi.gmu.edu/CSI/applications.html Information about CSI and the Krasnow Institute is available at http://csi.gmu.edu/ and http://www.gmu.edu/departments/krasnow/ Application deadline for the fall 1999 term is March 1 1999. From zhang at salk.edu Mon Feb 8 18:24:28 1999 From: zhang at salk.edu (Kechen Zhang) Date: Mon, 8 Feb 1999 15:24:28 -0800 (PST) Subject: Preprint available Message-ID: The following paper is available in PostScript form: Journal of Neuroscience, in press A theory of geometric constraints on neural activity for natural three-dimensional movement Kechen Zhang and Terrence J. Sejnowski Howard Hughes Medical Institute, The Salk Institute, and University of California, San Diego Abstract: Although the orientation of an arm in space or the static view of an object may be represented by a population of neurons in complex ways, how these variables change with movement often follows simple linear rules, reflecting the underlying geometric constraints in the physical world. A theoretical analysis is presented for how such constraints affect the average firing rates of sensory and motor neurons during natural movements with low degrees of freedom, such as a limb movement and rigid object motion. When applied to non-rigid reaching arm movements, the linear theory accounts for cosine directional tuning with linear speed modulation, predicts a curl-free spatial distribution of preferred directions, and also explains why the instantaneous motion of the hand can be recovered from the neural population activity. For three-dimensional motion of a rigid object, the theory predicts that, to a first approximation, the response of a sensory neuron should have a preferred translational direction and a preferred rotation axis in space, both with cosine tuning functions modulated multiplicatively by speed and angular speed, respectively. Some known tuning properties of motion-sensitive neurons follow as special cases. Acceleration tuning and nonlinear speed modulation are considered in an extension of the linear theory. This general approach provides a principled method to derive mechanism-insensitive neuronal properties by exploiting the inherently low dimensionality of natural movements. http://www.cnl.salk.edu/~zhang/objectwfigs.ps.Z 31 pages, compressed PostScript file about 2.0 Mb (7.7 Mb uncompressed) From Marc.VanHulle at med.kuleuven.ac.be Tue Feb 9 04:45:50 1999 From: Marc.VanHulle at med.kuleuven.ac.be (Marc Van Hulle) Date: Tue, 9 Feb 1999 10:45:50 +0100 Subject: NNSP'99 deadline extended to February 20 Message-ID: <199902090945.KAA21408@simone.neuro.kuleuven.ac.be> ************************************************ **** FINAL CALL FOR PAPERS **** **** **** **** Note extended submission deadline **** **** Submission deadline: February 20, 1999 **** ************************************************ 1999 IEEE Workshop on Neural Networks for Signal Processing August 23-25, 1999, Madison, Wisconsin NNSP'99 homepage: http://eivind.imm.dtu.dk/nnsp99 Thanks to the sponsorship of IEEE Signal Processing Society the ninth of a series of IEEE workshops on Neural Networks for Signal Processing will be held at the Concourse Hotel, Madison, Wisconsin, USA. The workshop will feature keynote addresses, technical presentations, panel discussions and special sessions. Papers are solicited for, but not limited to, the following areas: Paradigms: Artificial neural networks, support vector machines, Markov models, graphical models, dynamical systems, evolutionary computation, nonlinear signal processing, and wavelets. Application Areas: Image/speech/multimedia processing, intelligent human computer interfaces, intelligent agents, blind source separation, OCR, robotics, adaptive filtering, communications, sensors, system identification, issues related to RWC, and other general signal processing and pattern recognition. Theories: Generalization, design algorithms, optimization, parameter estimation, and network architectures. Implementations: Parallel and distributed implementation, hardware design, and other general implementation technologies. SPECIAL SESSIONS The workshop features special sessions on * Support vector machines * Intelligent human computer interfaces PAPER SUBMISSION PROCEDURE Prospective authors are invited to submit extended summaries of no more than 6 pages using the electronic submission procedure described at the workshop homepage http://eivind.imm.dtu.dk/nnsp99 Exceptionally, in case of mail submission, prepare 5 copies of extended summaries of no more than 6 pages. The top of the first page of the summary should include a title, authors' names, affiliations, address, telephone and fax numbers and email address. Please send paper submissions to: Jan Larsen NNSP'99, Department of Mathematical Modelling, Building 321 Technical University of Denmark DK-2800 Lyngby, Denmark Camera-ready full papers of accepted proposals will be published in a hard-bound volume by IEEE and distributed at the workshop. SCHEDULE Submission of extended summary: February 20, 1999 Notification of acceptance: March 31, 1999 Submission of photo-ready accepted paper: April 29, 1999 Advanced registration, before: June 30, 1999 ORGANIZATION General Chair Yu Hen HU University of Wisonsin-Madison email: hu at ece.wisc.edu Finance Chair Tulay ADALI University of Maryland Baltimore County email: adali at umbc.edu Proceedings Chair Elizabeth J. WILSON Raytheon Co. email: bwilson at ed.ray.com Proceedings Co-Chair Scott C. DOUGLAS Southern Methodist University email: douglas at seas.smu.edu Publicity Chair Marc van HULLE Katholieke Universiteit Leuven email: marc at neuro.kuleuven.ac.be From Michael_Tarr at brown.edu Tue Feb 9 11:36:28 1999 From: Michael_Tarr at brown.edu (Michael J. Tarr) Date: Tue, 9 Feb 1999 11:36:28 -0500 Subject: No subject Message-ID: For immediate posting: VISITING POSITION IN HUMAN PERCEPTION/COGNITION, BROWN UNIVERSITY. The Department of Cognitive and Linguistic Sciences at Brown University invites applications for a three-year visiting assistant professor position, to begin July 1, 1999 (non-tenure-track). Candidates should have an active research program in the area of perception, action, visual cognition, and/or attention, and would be expected to teach three courses per year. All applicants must have received the Ph.D. degree or equivalent by the beginning of the appointment. The initial deadline for applications is March 15, 1999, but applications will be accepted after that time until the position is filled. Please send CV, recent publications, a cover letter describing teaching and research interests, and three letters of reference to: Perception/Cognition Search Committee, Department of Cognitive and Linguistic Sciences, Box 1978, Brown University, Providence, RI 02912. Brown is an Equal Opportunity/Affirmative Action employer. Women and minorities are especially encouraged to apply. From sschaal at usc.edu Tue Feb 9 12:10:49 1999 From: sschaal at usc.edu (Stefan Schaal) Date: Tue, 9 Feb 1999 09:10:49 -0800 Subject: Postdoc position in Computational Motor Control Message-ID: <199902091710.JAA15291@rubens.usc.edu.> Postdoctoral Position in Computational Motor Control ----------------------------------------------------- The Computational Learning and Motor Control Laboratory at the University of Southern California, which is part of the Kawato Dynamic Brain Project at ATR in Japan, is looking for an outstanding postdoctoral fellow in Computational Motor Control. The position is available starting from April 99, initially for one year, but can be renewed until Sept. 2001. Research topics of our group are strongly interdisciplinary. On the one hand, we are interested in developing formal theories of how the mammalian brain solves problems of sensory motor control. On the other hand, we also try to test these theories with humanoid robots, ranging from human-arm-like robots up to a full-body humanoid with a complete oculomotor and vestibular system. We also employ behavioral experiments to compare human performance against robot performance and our developed theories. A strong emphasis of our research lies on developing statistical learning mechanisms for the self-organization of motor control, for instance, applied to the learning of internal models, the self-organization of internal representations, or high-level topics like imitation learning. More information can be found at: http://www-slab.usc.edu or http://www.erato.atr.co.jp The ideal candidate for this position has a strong background in statistical learning and neural networks, in computational neuroscience of motor control, in behavioral motor control, as well as in dealing with experimental and computer equipment. The candidate is expected to independently lead research projects, supervise students, and interact with our Japan-based part of the project through visits to Japan. The postdoctoral position allows a great deal of flexibility concerning the research topic as long as it fits within the broad range of our group's interests in computational motor control. Before sending any formal applications, including a CV and at least 3 letters of recommendation, send a (brief!!) email letter of intent to sschaal at usc.edu (Dr. Stefan Schaal), briefly describing your past and current research interests, and how you would imagine to integrate in our projects. Stefan Schaal Assistant Professor Computer Science and Neuroscience University of Southern California, HNB 103 Los Angeles, CA 90089-2520 & Head, Computational Learning Group Kawato Dynamic Brain Project (ERATO/JST) ATR, Japan phone: (213) 740 9418 fax: (213) 740 1510 From ckiw at dai.ed.ac.uk Tue Feb 9 13:07:17 1999 From: ckiw at dai.ed.ac.uk (Chris Williams) Date: Tue, 9 Feb 1999 18:07:17 +0000 (GMT) Subject: ICANN*99 Call for Workshop Proposals Message-ID: CALL FOR PROPOSALS ICANN*99 Post Conference Workshops 11 September 1999 Edinburgh, Scotland Workshops on current topics in the fields of research covered by ICANN99 will be held on Saturday 11 September 1999 immediately after the regular ICANN conference program. Proposals by qualified individuals interested in chairing one of these workshops are solicited. Topics should fall within the broadly defined area of neural networks research, including: Theory and Algorithms; Neurobiology and Computational Neuroscience; Cognitive Modelling; Industrial, Commercial and Medical Applications; Hardware and Neuromorphic Engineering; Control, Robotics and Adaptive Behaviour. The goal of the workshops is to provide an informal forum for researchers to discuss important issues of current interest. The workshops will be held on Saturday 11th September, the day following the conference, making it convenient for ICANN participants to stay in Edinburgh over a Saturday night and thus be eligible for cheaper flights. It is anticipated that the registration fee for the workshops will be nominal or perhaps zero. Please note that there is no funding available for the workshops. Focussed open and/or controversial issues are encouraged and preferred as workshop topics. Representation of alternative viewpoints and panel-style discussions are particularly encouraged. Workshop organizers will have responsibilities including: 1) coordinating workshop participation and content, which could involve, for example, arranging short informal presentations by experts working in an area, arranging for expert commentators to sit on a discussion panel and formulating a set of discussion topics, etc. 2) moderating or leading the discussion and reporting its high points, findings, and conclusions to the group during evening plenary sessions 3) writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. This call for proposals can be found at http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/icann/ and further information will be added as it becomes available. Submission Instructions ----------------------- Interested parties should submit via e-mail a short proposal for a workshop of interest by 29 March, 1999 to Dr Chris Williams (ckiw at dai.ed.ac.uk), Institute for Adaptive and Neural Computation, University of Edinburgh. Proposals should include a title, a description of what the workshop is to address and accomplish, the planned format (mini-conference, panel discussion, or group discussion, combinations of the above, etc), and the proposed number of speakers. Where possible, please also indicate potential invitees (particularly for panel discussions). We strongly encourage that the organizers reserve a significant portion of time for open discussion. 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. In addition, please send a brief resume of the prospective workshop chair, a list of publications, and evidence of scholarship in the field of interest. Submissions should include contact name, address, e-mail address, phone number and fax number if available. Proposals should be mailed electronically to ckiw at dai.ed.ac.uk. All proposals must be RECEIVED by March 29, 1999. If e-mail is unavailable, mail so as to arrive by the deadline to: ICANN*99 Workshops c/o Dr Chris Williams Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 5 Forrest Hill, Edinburgh EH1 2QL, Scotland, UK PROPOSALS MUST BE RECEIVED BY MARCH 29 1999 -Please Post- -------------------------------------------------------------------------- Chris Williams ckiw at dai.ed.ac.uk Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 5 Forrest Hill, Edinburgh EH1 2QL, Scotland, UK fax: +44 131 650 6899 tel: (direct) +44 131 651 1212 (department switchboard) +44 131 560 3090 http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/ From S.Holden at cs.ucl.ac.uk Thu Feb 11 08:07:50 1999 From: S.Holden at cs.ucl.ac.uk (Sean Holden) Date: Thu, 11 Feb 1999 13:07:50 +0000 Subject: Research Fellow post in Intelligent Data Analysis and Fusion Message-ID: <856.918738470@cs.ucl.ac.uk> University College London Department of Computer Science Research in Intelligent Data Analysis and Fusion A position is available for a Research Fellow to work in collaboration with SmithKline Beecham, Glaxo-Wellcome, Unilever and Zeneca on an EPSRC funded project selected by the NPL-Sira Faraday Partnership. The aim of the project is to develop a hybrid data mining, analysis and fusion system incorporating the most recent artificial intelligence, neural network, fuzzy logic, and genetic algorithm techniques. The research fellow will be responsible for development of a hybrid system tool, for day to day running of the project, for evaluating the performance of the system in collaboration with the above companies and NPL, and for overseeing technology transfer to Integral Systems Solutions Ltd, supplier of the Clementine data mining system. Applicants should have a strong background in artificial intelligence, computer science and the development of data mining systems. The Fellow will work in the Intelligent Systems Research Group with four research students on the project, one associated with each of the major collaborating companies. The Group has extensive experience in the development of hybrid intelligent systems, in particular for financial and commercial applications. Applicants must have good mathematical and computing skills with experience of data mining research or applications or have carried out research and development in a closely related area. The position is for up to three years. The Research Fellow will be on the University RA1A scale (17,869-21,331 pounds sterling, including London allowance, depending on experience and length of appointment). For more details of the post and project see: http://www.cs.ucl.ac.uk/staff/b.buxton/Faraday or e-mail at B.Buxton at cs.ucl.ac.uk. or tel: 0171-380-7294 Please send 3 copies of your CV (including the names of two referees) to: M Sutcliffe, Department of Computer science, University College London, Gower Street, WC1E 6BT. or e-mail: vacancy at cs.ucl.ac.uk. The closing date for applications is 3 March 1999. ************* From rcs at cogsci.ed.ac.uk Thu Feb 11 06:15:15 1999 From: rcs at cogsci.ed.ac.uk (rcs@cogsci.ed.ac.uk) Date: Thu, 11 Feb 1999 11:15:15 GMT Subject: MSc in Cognitive SCience and Natural Language Message-ID: <21892.199902111115@burns.cogsci.ed.ac.uk> ************************************************************************** ************************************************************************** MSc in "Cognitive Science and Natural Language" in the School of Cognitive Science, Division of Informatics, University of Edinburgh The new School of Cognitive Science, in the newly created Division of Informatics, takes over the teaching and research training activities of the former Centre for Cognitive Science. In an act of fundamental restructuring, the Centre for Cognitive Science and the Human Communication Research Centre, as well as the departments of Computer Science and Artificial Intelligence, have all merged to form the Division of Informatics. In very broad terms, Informatics is devoted to studying the structure, behaviour and design of computational, cognitive and communicating systems, both natural and artificial. TEACHING STAFF Mark Steedman Head of School computational linguistics, grammar formalisms, knowledge representation Chris Brew corpora, data intensive linguistics, language technology Jo Calder grammar formalisms, computational linguistics Bruce Graham computational neuroscience, neural networks Alexander Holt natural language semantics, computational linguistics Ewan Klein linguistic theory, phonology Alex Lascarides lexical and discourse processing, semantics, pragmatics Chris Mellish natural language processing and generation Paul Schweizer philosophical logic, philosophy of mind, philosophy of language Richard Shillcock psycholinguistics, cognitive modelling, cognitive neuropsychology Keith Stenning human memory, inference, connectionism Geraint Wiggins logic programming, intelligent music systems Roberto Zamparelli theoretical linguistics MSc POSTGRADUATE PROGRAMME The School of Cognitive Science directly continues the educational heritage of the Centre for Cognitive Science, which established its postgraduate programme in 1978 and is widely recognized internationally as one of the best programmes of it's kind. The School of Cognitive Science offers a course of postgraduate study centred on language and cognition. STRUCTURE OF THE PROGRAMME The specific modules available vary slightly from year to year, but the following list of seven contributing disciplines provides an accurate indication of the general pattern: Cognitive Psychology Computational Linguistics Data Intensive Linguistics Formal Logic Logic Programming and Knowledge Representation Neural Computation Theoretical Linguistics The teaching is carried out over two ten-week terms, and many of the contributing subjects have both first and second term modules. Students are expected to take four modules per term for assessment. Between May and September, MSc students work on a dissertation or project. In addition to the first year taught course, there are a number of advanced courses available within the Division of Informatics for students working beyond the MSc level. Also, there is a rich diversity of research workshops and working groups in which PhD students are active participants. STUDYING IN EDINBURGH Edinburgh contains the largest concentration of expertise in Artificial Intelligence and Natural Language Processing in Europe. Students have access to that expertise, to Edinburgh's large copyright libraries, and within Cognitive Science, to a substantial offprint library. The School possesses extensive computing facilities based on a network of Sun workstations and Apple Macintoshes; access to Edinburgh's concurrent supercomputer and other central computing services is easily arranged. Edinburgh, `Athens of the North', is a beautiful Georgian city. Home of the International Festival, it is endowed with green parklands and lies close to the wildscape of the Scottish Highlands. Edinburgh restaurants offer a wide range of ethnic cuisines, and there is a variety of theatres, galleries, museums and cinemas. Applicants typically have undergraduate degrees in one of the participating areas or an appropriate joint honours degree. UK and EU students following the MSc and PhD courses are eligible to apply for studentships. SCS will advise all students concerning funding possibilities. SCS attracts studentships from a variety of UK and non-UK funding bodies. Contact: Mrs Judith Gordon or Miss E Kerse MSc Admissions PhD Admissions Division of Informatics Division of Informatics University of Edinburgh University of Edinburgh 5 Forrest Hill James Clerk Maxwell Bldg Edinburgh EH1 2QL Kings Buildings UK Edinburgh EH9 3JZ UK email: msc-admissions at inf.ed.ac.uk phd-admissions at inf.ed.ac.uk FOR DETAILED UP TO DATE INFORMATION ON THE SCHOOL OF COGNITIVE SCIENCE h t t p : / / w w w . c o g s c i . e d . a c . u k / ************************************************************************** ************************************************************************** From ASJagath at ntu.edu.sg Thu Feb 11 20:50:52 1999 From: ASJagath at ntu.edu.sg (Jagath C Rajapakse (Dr)) Date: Fri, 12 Feb 1999 09:50:52 +0800 Subject: Postdoc position available in brain image analysis Message-ID: <6665AC0C667ED11186E308002BB487E10366D0F7@exchange2> Postdoctoral Position Available in Brain Image Analysis A postdoctoral research fellow position is available to work under the supervision of Dr. Jagath C. Rajapakse in the School of Applied Science, Nanyang Technological University, Singapore. The fellow will involve in a joint research project between the school and Singapore General Hospital investigating analysis of functional MR time-series and segmentation of neuroanatomical structures in structural MR head scans. Knowledge in image processing and experience in programming in C++ are required. Candidate should have a Ph.D. in a related field. For further information, contact: Dr. Jagath Rajapakse, School of Applied Science, Nanyang Technological University, N4 Nanyang Avenue, Singapore 639798. Email: asjagath at ntu.edu.sg, Phone: +65 790 5802, Fax: +65 792 6559 From stefan.wermter at sunderland.ac.uk Fri Feb 12 13:46:28 1999 From: stefan.wermter at sunderland.ac.uk (Stefan Wermter) Date: Fri, 12 Feb 1999 18:46:28 +0000 Subject: Job: Research Network Assistant Message-ID: <36C47704.712E6D2F@sunderland.ac.uk> Part-time Research Network Assistant Applications are invited for a two year research network assistant position. You will support a new international research network in the areas of neural networks and computational neuroscience in the framework of the EPSRC project EmerNet. The exciting new project examines the possible benefits of neuroscience and brain architectures for computational neural networks in the future. The candidate will play a key role in the support of this international research network including set up of web server for the network, restricted web programming, design and maintenance of web pages, collection of material in computational neuroscience, organisational support and administration for international workshops and technical writing. Candidates should have a degree in a Computing/Neurobiology/ Neuroscience discipline and an interest in artificial neural networks and computational neuroscience. Good communication and technical writing skills are required and some software engineering experience is an advantage. There will be a possibility to register for PhD part-time if so desired. 18.5 hours per week 12,733 - 14,222 (pro-rata) Informal enquiries welcomed by Professor Dr Stefan Wermter, e-mail Stefan.Wermter at sunderland.ac.uk Ref No: CETS48 Closing Date: 25/2/99 Application forms and further particulars are available by telephoning extensions 3578, 2429, 2055, or 2425 or E-Mail employee.recruitment at sunderland.ac.uk quoting appropriate reference number. ******************************************** Professor Stefan Wermter Research Chair in Intelligent Systems University of Sunderland Centre of Informatics School of Computing, Engineering and Technology St Peters Way Sunderland SR6 0DD United Kingdom phone: +44 191 515 3279 fax: +44 191 515 2781 email: stefan.wermter at sunderland.ac.uk http://www.his.sunderland.ac.uk/ http://osiris.sunderland.ac.uk/~cs0stw/ ******************************************** From Ronan.Reilly at ucd.ie Fri Feb 12 22:12:40 1999 From: Ronan.Reilly at ucd.ie (Ronan Reilly) Date: Fri, 12 Feb 1999 19:12:40 -0800 Subject: Call for submissions - Cognitive Science Conference - University College Dublin Message-ID: <009301be56fe$b947b980$6a042b89@numwaan.ucd.ie> CALL FOR SUBMISSIONS "Cognitive Science for the New Millennium" University College Dublin May 16-17, 1999 http://www.cs.ucd.ie/cogsci Guest Speakers James L. McClelland, Carnegie-Mellon University Paul Smolensky, Johns Hopkins University To mark the launch in the academic year 1999/2000 of the MA/MSc programme in Cognitive Science at University College Dublin, a conference is being organised to celebrate and evaluate the achievements of the discipline. Conference submissions are invited that deal with all aspects of cognitive science. Submissions are especially encouraged that focus on an evaluation of its achievements and/or on an exploration of promising future research directions. Submissions should comprise a one page extended abstract. The conference will have a single stream, involving just oral presentations. We strongly encourage submissions by e-mail as ASCII or postscript attachments sent to Ronan.Reilly at ucd.ie , with subject "cogsci submission". Hardcopy submissions should include three copies of the abstract and be sent to the address below. All communications regarding submissions will be by e-mail, so provision of an e-mail address is essential. All submissions must be received on or before the 31st of March. For more details see the conference web page at http://www.cs.ucd.ie/cogsci. _________________________________ Ronan Reilly, PhD Department of Computer Science University College Dublin Belfield Dublin 4 voice: +353-1-706 2475 fax: +353-1-269 7262 email: Ronan.Reilly at ucd.ie web: http://www.cs.ucd.ie/staff/rreilly From mmp at ai.mit.edu Sat Feb 13 18:12:27 1999 From: mmp at ai.mit.edu (Marina Meila) Date: Sat, 13 Feb 99 18:12:27 EST Subject: PhD Thesis: "Learning with MIXTURES OF TREES" - Marina Meila - MIT Message-ID: <9902132312.AA13085@sinaia.mit.edu> Dear connectionists, my thesis on learning tree graphical models and mixtures thereof is now available via my web page: http://www.ai.mit.edu/~mmp or directly from ftp://ftp.ai.mit.edu/pub/mmp/thesis/thesis.ps.Z The thesis presents: * Density estimation in multidimensional domains and graphical models * Tree distributions and their properties * Mixtures of trees: definition, properties, an efficient learning algorithm * Learning mixtures of trees in the Bayesian framework * Very fast tree learning algorithms for sparse discrete data * A new approach to hidden variable discovery, an empirical method for estimating the description length of a mixture model, a novel independence test based on large deviation theory * Experiments: density estimation and classification with mixtures of trees Abstract and Table of Contents follow: --------------------------------------------------------------------------- Abstract One of the challenges of density estimation as it is used in machine learning is that usually the data are multivariate and often the dimensionality is large. Operating with joint distributions over multidimensional domains raises specific problems that are not encountered in the univariate case. Graphical models are representations of joint densities that are specifically tailored to address these problems. They take advantage of the (conditional) independencies between subsets of variables in the domain which they represent by means of a graph. When the graph is sparse, graphical models provide an excellent support for human intuition and allow for efficient inference algorithms. However, learning the underlying dependence graph from data is generally NP-hard. The purpose of this thesis is to propose and to study a class of models that admits tractable inference and learning algorithms yet is rich enough for practical applications. This class is the class of mixtures of trees models. Mixtures of trees inherit the excellent computational properties of tree distributions (themselves a subset of graphical models) but combine several trees in order to augment their modeling power, thereby going beyond the standard graphical model framework. The thesis demonstrates the performance of the mixture of trees in density estimation and classification tasks. In the same time it deepens the understanding of the properties of the tree distribution as a multivariate density model. Among others, it shows that the tree classifier implements an implicit variable selection mechanism. Learning mixtures of trees from data is a central subject of this thesis. The learning algorithm that is introduced here is based on the the EM and the Minimum Weight Spanning Tree algorithms and is quadratic in the dimension of the domain. This algorithm can serve as a tool for discovering hidden variables in a special but important class of models where, conditioned on the hidden variable, the dependencies between the observed variables become sparse. Finally, it is shown that in the case of sparse discrete data, the original learning algorithm can be transformed in an algorithm that is jointly subquadratic and that in simulations achieves speedups factors of up to a thousand. ---------------------------------------------------------------------------- Learning with Mixtures of Trees Marina Meila-Predoviciu Table of Contents * Cover, etc (postscript, 14 pages) o Abstract o Acknowledgements o Contents o List of Figures o List of Tables o Index of Algorithms * Introduction (postscript, 12 pages) o Density estimation in multidimensional domains o Introduction by example o Graphical models of conditional independence + Examples of established belief network classes + Advantages of graphical models + Structure learning in belief networks + Inference and decomposable models o Why, what and where? Goal, contributions and road map of the thesis + Contributions + The mixture of trees model + An efficient learning algorithm + An accelerated learning algorithm for sparse data + A top-down approach to hidden variable discovery + A road map for the reader * Trees and their properties (postscript, 14 pages) o Tree distributions o Inference, sampling and marginalization in a tree distribution + Inference + Marginalization + Sampling o Learning trees in the Maximum Likelihood framework + Problem formulation + Fitting a tree to a distribution + Solving the ML learning problem o Representation capabilities o Appendix: The Junction Tree algorithm for trees * Mixtures of trees (postscript, 12 pages) o Representation power of mixtures of trees o Basic operations with mixtures of trees + Marginalization + Inference + Sampling o Learning mixtures of trees in the ML framework + The basic algorithm + Running time and storage requirements + Learning mixtures of trees with shared structure + Remarks on the learning algorithms o Summary and related work * Learning mixtures of trees in the Bayesian framework (postscript, 9 pages) o MAP estimation by the EM algorithm o Decomposable priors for tree distributions + Decomposable priors over tree structures + Priors for tree parameters: the Dirichlet prior + The Dirichlet prior in natural coordinates + Dirichlet priors for trees and mixtures * Accelerating the tree learning algorithm (postscript, 22 pages) o Introduction o Assumptions o Accelerated CL algorithms + First idea: Comparing mutual informations between binary variables + Second idea: computing cooccurrences in a bipartite graph data representation + Putting it all together: the aCL-I algorithm and its data structures + Time and storage requirements + The aCL-II algorithm + Time and memory requirements for aCL-II o Generalization to discrete variables of arbitrary arity + Computing cooccurrences + Presorting mutual informations + A ``chain rule'' expression for the entropy of a discrete variable + The mutual information of two non-cooccurring variables o Using the aCL algorithms with EM o Decomposable priors and the aCL algorithm o Experiments o Concluding remarks o Appendix: Bounding the number of lists * An Approach to Hidden variable discovery (postscript, 21 pages) o Structure learning paradigms o The problem of variable partitioning o The tree H model o Variable partitioning in the general case + Outline of the procedure + Defining structure as simple explanation o Experiments + Experimental procedure + Experiments with tree H models + General H models o Approximating the description length of a model + Encoding a multinomial distribution o Model validation by independence testing + An alternative independence test + A threshold for mixtures + Validating graphical models with hidden variables o Discussion * Experimental results (postscript, 17 pages) o Recovering the structure o Density estimation experiments + Digits and digit pairs images + The ALARM network and data set o Classification with mixtures of trees + Using a mixture of trees as a classifier + The AUSTRALIAN data set + The MUSHROOM data set + The SPLICE data set. Classification and structure discovery + The single tree classifier as an automatic feature selector * Conclusion (postscript, 2 pages) * References (postscript, 5 pages) or (html) ---------------------------------------------------------------------------- From horn at neuron.tau.ac.il Sun Feb 14 03:26:32 1999 From: horn at neuron.tau.ac.il (David Horn) Date: Sun, 14 Feb 1999 10:26:32 +0200 (IST) Subject: Position in Functional Brain Imaging at Tel Aviv University Message-ID: TEL AVIV UNIVERSITY SENIOR ACADEMIC POSITION IN FUNCTIONAL BRAIN IMAGIMG ADAMS SUPER CENTER FOR BRAIN STUDIES Applications are invited for a senior academic position in functional brain imaging. The successful applicant should have experience in fMRI and will be expected to lead a research program in this field in collaboration with different disciplines throughout Tel Aviv University. The appointment procedure shall be according to the nomination regulations of Tel Aviv University. Departmental affiliation will be decided according to the academic background of the appointee. Applications including a CV, list of publications and a short summary of research interests should be sent by April 15th, 1999 to: Prof. David Horn, Adams Super Center for Brain Studies, Kaplun Bldg. room #428, Tel Aviv University, Tel Aviv, 69978, Israel. (Fax: ++972-3-6407932; e-mail: brainctr at neuron.tau.ac.il). Candidates should arrange to have three letters of reference transmitted directly to the above address. From esann at dice.ucl.ac.be Mon Feb 15 06:36:20 1999 From: esann at dice.ucl.ac.be (ESANN) Date: Mon, 15 Feb 1999 12:36:20 +0100 Subject: ESANN'99 - European Symposium on Artificial Neural Networks Message-ID: <002201be58d7$68aa3270$6f1f6882@natacha.dice.ucl.ac.be> ---------------------------------------------------- | | | ESANN'99 | | | | 7th European Symposium | | on Artificial Neural Networks | | | | Bruges (Belgium) - April 21-22-23, 1999 | | | | Preliminary programme | ---------------------------------------------------- The preliminary programme of the ESANN 99 conference is now available on the Web: http://www.dice.ucl.ac.be/esann For those of you who maintain WWW pages including lists of related ANN sites: we would appreciate if you could add the above URL to your list; thank you very much! We try as much as possible to avoid multiple sendings of this call for papers; however please apologize if you receive this e-mail twice, despite our precautions. For 7 years the ESANN conference has become a major event in the field of neural computation. ESANN is a human-size conference focusing on fundamental aspects of artificial neural networks (theory, models, algorithms, links with statistics, data analysis, biological background,...). The programme of the conference can be found at the URL http://www.dice.ucl.ac.be/esann, together with practical information about the conference venue, registration,... Other information can be obtained by sending an e-mail to esann at dice.ucl.ac.be . ===================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - Microelectronics Laboratory 3, pl. du Levant - B-1348 Louvain-la-Neuve - Belgium tel: +32 10 47 25 51 - fax: + 32 10 47 25 98 mailto:esann at dice.ucl.ac.be * Conference secretariat D facto conference services 27 rue du Laekenveld - B-1080 Brussels - Belgium tel: + 32 2 420 37 57 - fax: + 32 2 420 02 55 mailto:esann at dice.ucl.ac.be ===================================================== From j-patton at nwu.edu Mon Feb 15 13:55:52 1999 From: j-patton at nwu.edu (Jim Patton) Date: Mon, 15 Feb 1999 10:55:52 -0800 Subject: Conference Announcement & Call for Papers Message-ID: <4.1.19990215104618.00bcc3c0@merle.acns.nwu.edu> 3rd IMACS/IEEE International Multiconference on Circuits, Systems, Communications and Computers Athens, Greece, JULY 4-8, 1999 Special Session on Control in Biological Systems Session Organizer: Dr. Kamran Iqbal Biological systems comprise all living organism, in particular the vertebrates. There has been a growing interest in the study of such systems since the early work of Sherington (The integrative action of the nervous system,1906) . Control in biological systems takes many forms, e.g., control of voluntary movements, control of gait and posture, control of muscle tension and joint stiffness, control of temperature, control of blood pressure, control of respiratory system, etc. Biological control is exerted through commands originating in the central nervous system, and is assisted by spinal reflexes and a low level motor servo. The neuro-physiological controller can be termed as autonomous, self-organizing, self-evolving, and hierarchical. Biological systems are distinguished by their complex and diverse nature. Presence of nonlinear dynamics, multiple degrees-of-freedom, inter-segmental couplings, controller and actuator redundancies, multiple levels of feedback, finite feedback delays, and the distributed nature of parameters makes modelling of biological systems a daunting task. The complexity inherent in biological control systems has long challenged the researchers and often precluded the use of many a powerful tools otherwise available to control theorists. New and exciting additions to systems theory have often evolved from study of biological systems, e.g., artificial neural networks. The proposed session aims to explore the control issues relevant to the biological systems. A non-exhaustive list of topics includes the following: physiological and neuro-muscular system modeling, musculo-skeletal models, biokinematics, biodynamics, biomechanics, motor control, motor learning, neurological control of muscle activity, skeltomotor reflexes, control of limb movement, control of posture and locomotion, control of skilled movements, control of manipulative abilities, control of blood pressure, respiratory/ventilatory control, biomedical applications of automatic control, functional electrical and functional neuromuscular stimulation, evolutionary approaches in biological systems, trajectory control of movements, identification and computer simulation of biological systems. Instructions to the authors: Please submit the following items by March 15, 1999: #1. A one page abstract via e-mail to: k-iqbal at nwu.edu (See Sample) #2. PDF(*.pdf) & MS Word (*.doc) file(s) for the full paper(s) via email (See Sample) #3. The CONSENT TO PUBLISH & TRANSFER OF COPYRIGHT FORM completed and signed (Please download and print the Form and mail to address below): Please Note: Only original papers, written in English, which have not previously been published elsewhere will be accepted. Overview papers which will be very well organized and give an extended survey of results in their topic are particularly welcome. For all other questions please contact k-iqbal at nwu.edu or write to: ***************************************************************** Dr. Kamran Iqbal 645 N. Michigan Ave Suite 1100 Chicago, IL 60611 Ph: 312-503-6801 ***************************************************************** Benefits of Contributing to the IMAC/IEEE MultiConference *Have your paper published in the Proceedings that will be distributed to all conference attendees (including you!) *Have your paper published in the Luxurious Books of World Scientific (independently on the Proceedings) *Obtain international recognition in front of the IEEE - IMACS audience *Exchange knowledge with qualified industry professionals *Enhance your CV with two independent Publications (one in the Proceedings, one in the Post-Conference Luxurious Books) *Gain recognition as a leader from your peers in the Circuits/Systems/Communications/Computers Engineering *Have the opportunity to visit Greece and combine the conference with your summer vacations in sun-drenched Greek islands. From barba at cvs.rochester.edu Fri Feb 12 13:03:24 1999 From: barba at cvs.rochester.edu (Barbara Arnold) Date: Fri, 12 Feb 1999 13:03:24 -0500 Subject: Summer Workshop on Perception, Action and Cognition (for undergrads) Message-ID: The University of Rochester, Center for Visual Science invites undergraduates to participate in our 1999 Summer Workshop "Perception, Action and Cognition" to be held August 6-8, 1999. The workshop will include lectures, and poster sessions, and will offer an interactive forum between participants and our faculty. The workshop will focus on technological innovations in visual perception and cognition, sensory-motor coordination, computational modeling of neuronal circuits, and cortical physiology and development. Hands-on demos will introduce participants to the diverse methodology used in our laboratories. We will provide fellowships for travel and living expenses for over 20 students who demonstrate an interest in vision and cognitive neuroscience. We encourage applications from students in their junior year. Contact: Barbara Arnold, Center for Visual Science, Uniiv of Rochester, Rochester, NY 14627, 716-275-2459, Barba at cvs.rochester.edu ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Barbara N. Arnold Administrator email: barba at cvs.rochester.edu Center for Visual Science phone: 716 275 8659 University of Rochester fax: 716 271 3043 Meliora Hall 274 Rochester NY 14627-0270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From tom at cse.unsw.edu.au Wed Feb 17 00:26:31 1999 From: tom at cse.unsw.edu.au (Tom Gedeon) Date: Wed, 17 Feb 1999 16:26:31 +1100 (EST) Subject: ICONIP'99 Call for Papers Message-ID: <990217052631.12097@cse.unsw.edu.au> Subject: ICONIP/ANZIIS/ANNES/ACNN'99 Call for papers ************************************ * * * ICONIP'99 * * jointly with * * ANZIIS'99 and ANNES'99 and ACNN'99 * * * ************************************ The 6th International Conference on Neural Information Processing-- jointly with the 7th Australian and New Zealand International Conference on Intelligent Information Processing Systems, and the 5th New Zealand International Conference on Artificial Neural Networks and Expert Systems, and the 11th Australian Conference on Neural Networks. ********************** * * * 16-20 November, 1999 * * Perth, Australia * * * ********************** In 1999, the annual conference of the Asian Pacific Neural Network Assembly, ICONIP'99, will be held jointly with three other major international conferences in the Asian Pacific Region, from 16 to 20 November 1999 in Perth (note the new venue!), Western Australia. The joint conference will have three parallel streams: ************************************************************ * * * Stream1: Biological Neural Information Processing * * Stream2: Neural Information Processing and Soft Computing * * Stream3: Applied Techniques & Intelligent Info. Systems * * * ************************************************************ TOPICS OF INTEREST Stream1: Biological Neural Information Processing Neurobiological systems Cognition Cognitive models of the brain Dynamical modelling, chaotic processes in the brain Brain computers, biological computers Consciousness, awareness, attention Adaptive biological systems Modelling emotions Perception, vision Learning languages Stream2: Neural Information Processing and Soft Computing Soft computing--paradigms, methods, tools Artificial neural network models, architectures and algorithms Evolutionary programming and genetic algorithms Fuzzy systems Optimisation Expert systems Artificial life Adaptive systems Approximate reasoning Hybrid and ensemble systems Distributed AI systems, agent-based systems Machine learning, data mining and intelligent databases Probabilistic and statistical methods Stream3: Applied techniques and Intelligent information systems Intelligent information retrieval systems Pattern recognition and image processing Speech recognition and language processing Human-computer interfaces, web computing Robotics and mechatronics Information engineering Time-series prediction Intelligent control Virtual reality Application of intelligent information technologies in: engineering, process industries, law, finance and business, medicine and sport, geographic information systems, geology, data mining, environmental protection, ... ***************************************************************************** HONORARY CHAIR Shun-Ichi Amari, Tokyo University GENERAL CONFERENCE CHAIR , Tamas (Tom) Gedeon, University of New South Wales GENERAL CONFERENCE CO-CHAIR Yianni Attikiouzel, University of Western Australia Marwan Jabri, University of Sydney Nikola Kasabov, University of Otago PROGRAM COMMITEE CO-CHAIRS K Fukushima (jp), T Gedeon, S Halgamuge (au), N Kasabov (nz), D Nauck (de) INTERNATIONAL PROGRAM COMMITTEE (TENTATIVE) T Adams (fj), D Aha (us), D Alahakoon (au), I Aleksander (uk), S Amari (jp), J Andreae (nz), Y Attikiouzel (au), J Austin (uk), J Baldwin (jp), P Bartlett (au), A Bastian, J Bezdek (us), R Bogner (au), B Bouchon-Meunier (fr), B Bruegge (de), A Bulsara (us), T Caelli (au), T Chen (cn), V Cherkassky (us), S Cho, S Cho (kr), A Cichocki (jp), I Cloete (za), R Coggins (au), G Coghill (nz), G Deboeck (us), J Diederich, V Dimitrov (au), L Ding (sg), T Downs (au), W Duch (pl), W Dunin-Barkowski (su), R Eckmiller (de), F Esteva (es), E Falkenauer (be), M Fedrizzi (it), B Flower (au), D Fogel (us), W Friedrich (nz), K Fukushima, T Furuhashi (jp), T Gedeon (au), M Glesner (de), J Grantner (us), S Halgamuge (au), M Hassoun (us), H Hellendoorn (nl), T Hendtlass (au), K Hirota (jp), A Hoffmann, J Hughes (au), M Ishikawa (jp), M Jabri, I Jagielska, L Jain (au), M Jamshidi (us), A Jennings (au), S Jones (uk), J Kacprzyk (pl), V Karri (au), N Kasabov (nz), O Katai, H Kawakami (jp), O Kaynak (tr), L Koczy, S Kovacs (hu), A Kowalczyk (au), W Kurutach (th), R Kozma (nz), R Kruse (de), D Lakov (bg), S Lee (kr), B Lees, D Levy (au), M Lim (sg), P Maheshwari (au), R Marks (us), F Masulli (it), G Matsumoto (jp), D McMichael (au), J Miller (nz), S Mitra (us), M Mizumoto (jp), N Morgan (us), M Mukaidono, T Musha, T Nagano (jp), D Nauck (de), M Negnevitsky (au), M Negoita (us), C Newton, C Latimer (au), R O'Shea (nz), E Oja (fi), Y Okabe (jp), N Pal, S Pal (in), M Palaniswami (au), G Pasi (it), L Patnaik (in), M Paulin (nz), D Pham (uk), A Ralescu (tr), B Reusch (de), L Reznik (au), D Ruan (be), I Rudas (hu), P Sallis (nz), E Sanchez (fr), D Saffen, Y Sato (jp), B Schuermann (de), O Simula (fi), U Srinivasan (au), S Sugiyama (jp), R Sun, H Szu (us), M Takacs (yu), H Takagi (jp), J Taylor (uk), A Topchy (ru), P Treleavan (uk), N Tschiold-Guerman (ch), A Tsoi (au), Y Tsujimura, M Tsukada, E Uchino, S Usui (jp), J Herik (nl), V Vapnik, R Vemuri (us), N Vojdani, W Wahlster (de), A Waibel (us), P Wallis, J Wiles, W Wilson, P Wong (au), L Xu (hk), T Yamakawa, T Yanaru (jp), X Yao (au), D Yun, L Zadeh, J Zurada (us), ORGANISING COMMITTEE CO-CHAIRS: Patrick Wong (Chair), Tom Gedeon, Graham Mann CONFERENCE ORGANISER Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au CALL FOR PAPERS AND FORMAT OF SUBMISSION Papers must be received by 16 April 1999. They will be reviewed by senior researchers in the field and the authors will be informed about the decision of the review process by 30 June 1999. The accepted papers must be submitted in a camera-ready format by 31 August. All accepted papers will be published by IEEE Computer Society Press. As the conference is a multi-disciplinary meeting the papers are required to be comprehensible to a wider rather than to a very specialised audience. Papers will be presented at the conference either in an oral or in a poster session. Please submit five copies of the paper written in English on A4-format white paper with 2.5 cm margins on all four sides, in two column format, on not more than 6 pages, single-spaced, in Times or similar font of 10 points, and printed on one side of the page only. Centred at the top of the first page should be the complete title, author(s), mailing and e-mailing addresses, followed by an abstract and the text. In the covering letter the stream and the topic of the paper according to the list above should be indicated. The IEEE Transaction journals LaTex article style can be used. SPECIAL ISSUES OF JOURNALS AND EDITED VOLUMES Selected papers will be published in special issues of scientific journals. The organising committee is looking for publications of edited volumes which include chapters covering the conference topics written by invited conference participants. TUTORIALS (16 November) Conference tutorials will be organized to introduce the basics of cognitive modelling, dynamical systems, neural networks, fuzzy systems, evolutionary programming, soft computing, expert systems, hybrid systems, and adaptive systems. Proposals for tutorials are due on 1 March 1999. EXHIBITION Companies and university research laboratories are encouraged to exhibit their developed or distributing software and hardware systems. STUDENT SESSION Postgraduate students are encouraged to submit papers to this session following the same formal requirements for paper submission. The submitted papers will be published in a separate brochure. VENUE (Perth) The Conference will be held at the Langley Hotel in Perth city, near the Opera House. The closing session will be held on Saturday, 20 November as a scenic bush location following an outdoor seafood barbecue. ACCOMMODATION Accommodation has been booked at the conference venue, and in the numerous Hotels nearby suitable for a range of budgets. Please see the web site for details after the paper deadline date. TRAVELLING To be added. POSTCONFERENCE EVENTS The ICONIP'99 post conference workshop and expo on "Future Directions for Intelligent Information Processing Systems" will be held in Dunedin, New Zealand from 22-23 November 1999. Please see the workshop homepage "http://divcom.otago.ac.nz/infosci/kel/iconip99-workshop.htm". An affiliated post-conference workshop is the third Australia-Japan joint Workshop on Intelligent and Evolutionary Systems will be held in Canberra, Australia from 22-23 November 1999. Please see the workshop homepage "http://www.cs.adfa.edu.au/conference/aj99/" IMPORTANT DATES Papers due: 16 April 1999 Proposals for tutorials: 1 March 1999 Notification of acceptance: 30 June 1999 Final camera-ready papers due: 31 August 1999 Registration of at least one author of a paper: 31 August 1999 Early registration: 31 August 1999 CONFERENCE CONTACTS, PAPER SUBMISSIONS, CONFERENCE INFORMATION, REGISTRATION FORMS Conference Secretariat ICONIP'99 Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au web: http://www.at.net.au/iconip99 ------------------------------------------------------------------------ ICONIP'99 jointly with ANZIIS'99 and ANNES'99 and ACNN'99 TENTATIVE REGISTRATION PLEASE PRINT Title:________________________________ Surname:______________________________ First Name:___________________________ Position:_____________________________ Organisation:_________________________ Department:___________________________ Address:______________________________ ______________________________________ City:_________________________________ Country:______________________________ Phone:________________________________ Fax:__________________________________ Email:________________________________ Yes/No. Will you attend the conference? Yes/No. Will you submit a paper? Yes/No. Will you attend the closing session on the cruise? Yes/No. Would you like any further information? Please send a copy of this completed form by E-mail to iconip99 at at.net.au or by post to: Conference Secretariat ICONIP'99 ------------------ Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983 Australia ------------------------------------------------------------------------ From m.niranjan at dcs.shef.ac.uk Wed Feb 17 05:24:46 1999 From: m.niranjan at dcs.shef.ac.uk (Mahesan Niranjan) Date: Wed, 17 Feb 1999 10:24:46 GMT Subject: JobJobJob Message-ID: <199902171024.KAA08190@bayes.dcs.shef.ac.uk> +++++Job++++++++Job+++++++++Job++++++++Job++++++++++Job++++++++Job++++++++ Department of Computer Science, The University of Sheffield, England. Post-Doctoral Research Assistant and PhD Studentship in Neural Networks and Machine Learning. Applications are invited for a post-doctoral research assistant and a PhD student to work on an EPSRC funded project: ``Critical Evaluation of Bayesian Methods in Neural Networks''. Both positions are for three years and available for immediate start. The age-linked salary for the RA is in the range 15,735 - 20,107 pounds per annum; standard EPSRC rules apply to the studentship. A longer description of the project can be found http://www.dcs.shef.ac.uk/~niranjan/jobs/rafp.html To apply, an application package should be obtained from: Personnel Department The University of Sheffield Firth Court, Sheffield S10 2TN http://www.shef.ac.uk/jobs; Phone: 44 144 222 1631 Closing date 01 March 1999 For informal inquiries and any specific questions on the research, contact Email: M.Niranjan at dcs.shef.ac.uk; Phone: 44 114 222 1805 +++++Job++++++++Job+++++++++Job++++++++Job++++++++++Job++++++++Job++++++++ From pj+ at andrew.cmu.edu Wed Feb 17 15:28:05 1999 From: pj+ at andrew.cmu.edu (Patsy J Mccarthy) Date: Wed, 17 Feb 1999 15:28:05 -0500 (EST) Subject: faculty position in Mathematical Sciences at CMU Message-ID: Please send the following job announcement to your mailing list. CARNEGIE MELLON UNIVERSITY Department of Mathematical Sciences The Department of Mathematical Sciences at Carnegie Mellon University seeks candidates with research interests in mathematical/computational biology. Qualified candidates at all levels are encouraged to apply. We are particularly interested in researchers with interests in neural modelling, brain function, and genome mathematics. A successful candidate for this position would be expected to interact with allied scientists in other centers and departments at Carnegie Mellon. Applicants should send a vita, list of publications, and a statement describing current and planned research. Candidates should also arrange to have at least three letters of recommendation sent to: Appointments Committee, Department of Mathematical Sciences, Carnegie Mellon University, Pittsburgh, PA 15213. Applications will be accepted until the position is filled. Carnegie Mellon University is an Affirmative Action/Equal Opportunity Employer. ============================================================================= Patsy McCarthy | /\ /\ Assistant to the Dept. Head | =( o o )= Dept. of Math Sciences, Carnegie Mellon Univ | / ~ \ WeH 6109, Ext. 82545 | ( | | ) ~^ ^~} From uwe.zimmer at gmd.gr.jp Wed Feb 17 23:53:27 1999 From: uwe.zimmer at gmd.gr.jp (Uwe R. Zimmer) Date: Thu, 18 Feb 1999 13:53:27 +0900 Subject: PostDoc Pos at GMD-Japan RL (Distributed Systems) Message-ID: <36CB9C72.9581E5DC@gmd.gr.jp> PostDoc-Pos-Announcement -------------------------------------------------------- Post-Doctoral Research Positions in Distributed Systems -------------------------------------------------------- Two new post-doctoral positions are open at GMD Japan Research Laboratory, Kitakyushu, Japan and will be filled at the earliest convenience. The new laboratory (started in November '98 with a team of 8 scientists and 4 support staff, where these first 8 positions should be filled until April '99) is based on long term cooperations with the Japanese research community and focuses on the robotics and the telecooperation research fields. Investigated topics (in the area of telecooperation) are: - Distributed systems - Awareness in remote or of global situations - Remote operation - Non-verbal communications - Coupled virtual environments Computer science aspects of distributed systems, several application layers (currently open for discussion) as well as connected psychological constraints and effects need to be considered simultaniously (but not necessarely by every individual researcher). Although we are an independent research group, there are of course close connections to GMD in Germany (German National Research Center for Information Technology - http://www.gmd.de/). GMD Japan Research Laboratory is participating in the currently established Gigabit-network project in Japan. If the above challenges rose your interest, please proceed to our expectations regarding the ideal candidate: - Ph.D. / doctoral degree in computer sciences, electrical engineering, physics, mathematics, psychology or related disciplines. - Experiences in experimenting in distributed environments - Theoretical foundations in mathematics, control, protocols, coding, or systems theory. - Interest in joining an international team of motivated researchers. Furthermore it is expected that the candidate evolves/introduces her/his own perspective on the topic, and pushes the goals of the whole group at the same time. Salary starts at 8 Mill. Yen per year depending on experience. Support concerning the transfer to the Japanese environment is provided. For any further information, and applications (including addresses of referees, two recent publications, and a letter of interest!) please contact: Uwe R. Zimmer (address below) P.S. preferred formats for applications: pdf, ps.z, or FrameMaker ___________________________________________ ____________________________| Dr. Uwe R. Zimmer ___| GMD - Japan Research Laboratory | AIM Building 8F, 3-8-1, | Asano, Kokurakita-ku | Kitakyushu-city, 802-0001, Japan | _________________________________________________________________| Voice: +81 93 512 1566 - Fax: +81 93 512 1588 | http://www.gmd.gr.jp/ | From baveja at research.att.com Thu Feb 18 09:18:53 1999 From: baveja at research.att.com (satinder singh baveja) Date: Thu, 18 Feb 1999 09:18:53 -0500 Subject: Extended Deadline for MLJ Special issue on Reinforcement Learning... Message-ID: <36CC214C.CD483A76@research.att.com> Hello All, Given that the UAI conference deadline is Feb 21 and Colt is March 1st, I am extending the submission deadline for the Machine Learning Journal Special Issue on Reinforcement Learning by 15 days to March 15. Several folks have written to me requesting this. If you are going to submit - an early e-mail submission of title and abstract (even if it changes a bit in the final submission) will be much appreciated. best wishes, satinder singh -------------- next part -------------- With apologies for multiple postings.... --------------------------------------------------------------------------- Machine Learning Journal Special Issue on REINFORCEMENT LEARNING Edited by Satinder Singh **Extended ** Submission Deadline: March 15, 1999 Reinforcement learning has become an exciting focus for research in machine learning and artificial intelligence. Some of this new excitement comes from a convergence of computational approaches to planning and learning in large-scale environments, some from accumulating empirical evidence that reinforcement learning algorithms can solve significant real-world problems, and some from the ever-increasing mathematical maturity of this field. In this special issue we would like to capture a snapshot of this recent excitement, and invite submission of papers describing new work in reinforcement learning. New algorithms, problem frameworks, theoretical results, and empirical results are all appropriate contributions. Submission Deadline: March 15, 1999 **** Extended deadline **** Expected publication Date: January, 2000 SUBMISSION INSTRUCTIONS *********************** Submissions will undergo the standard Machine Learning journal review process. Send one hard-copy of submissions to: Satinder Singh AT&T Shannon Lab Room A269 180 Park Avenue Florham Park, NJ 07932 USA Fax: (973) 360-8970 E-mail: baveja at research.att.com If possible, e-mail the abstract separately to baveja at research.att.com, even before it is final, so that reviewers can be allocated efficiently. Also mail five hard-copies of submitted papers to: Karen Cullen MACHINE LEARNING Editorial Office Kluwer Academic Publishers 101 Philip Drive Assinippi Park Norwell, MA 02061 USA phone: (617) 871-6300 E-mail: karen at world.std.com **** PLEASE INDICATE that your submission is for the REINFORCEMENT **** LEARNING SPECIAL issue. Submissions exceptionally longer than 24 pages when formatted according the journal's style (pointers to which are below) may be rejected without review. Note: Machine Learning is now accepting submission for FINAL copy in electronic form. There is a latex style file and a related files available via the url: http://www.wkap.nl/journalhome.htm/0885-6125 Authors are strongly encouraged to use these style files or the formatting instructions stated at the back of the Machine Learning journal for their submissions. From mel at lnc.usc.edu Thu Feb 18 20:37:31 1999 From: mel at lnc.usc.edu (Bartlett Mel) Date: Thu, 18 Feb 1999 17:37:31 -0800 Subject: Preprint: Why Have Dendrites? Message-ID: <36CCC05B.B503E48C@lnc.usc.edu> The following preprint is now available: ========================================================= "Why Have Dendrites? A Computational Perspective" Bartlett W. Mel Biomedical Engineering Department University of Southern California ----- To appear in "Dendrites". G. Stuart, N. Spruston, & M. Hausser, (Eds.), Oxford University Press, 1999. Download: http://lnc.usc.edu/abstracts/99.mel_dendrites.html 18 pages, 431K gzipped postscript ----- ABSTRACT In most neurons, dendrites receive the vast bulk of the cell's synaptic input. But why do dendrites exist? What benefits do they confer upon a neuron? What computing operations are carried out in dendritic trees which could not occur within an electrically compact neuron? In this chapter we consider evidence from both physiological and computer modeling studies suggesting that dendrites provide space for a large number of compartmentalized nonlinear synaptic interactions. We thus narrow our sights to consider only those forms of dendritic integration in which spatial de-localization of synaptic inputs is a necessary feature of the postsynaptic computation. We illustrate the power and generality of this view of dendritic function by enumerating several diverse ways in which intradendritic computations could contribute to known nonlinear receptive field properties in visual cortex. -- Bartlett W. Mel (213)740-0334, -3397(lab) Assistant Professor of Biomedical Engineering (213)740-0343 fax University of Southern California, OHE 500 mel at lnc.usc.edu, http://lnc.usc.edu US Mail: BME Department, MC 1451, USC, Los Angeles, CA 90089 Fedex: 3650 McClintock Ave, 500 Olin Hall, LA, CA 90089 From sima at uivt.cas.cz Fri Feb 19 04:07:53 1999 From: sima at uivt.cas.cz (Jiri Sima) Date: Fri, 19 Feb 1999 10:07:53 +0100 Subject: New report available Message-ID: <36CD29E9.2440408A@uivt.cas.cz> Dear Colleagues, the following report Sima, J., Orponen, P. A Continuous-Time Hopfield Net Simulation of Discrete Neural Networks. Technical report V-773, ICS CAS, Prague, January, 1999. is now available on-line at http://www.uivt.cas.cz/ics/reports.html Abstract: We investigate the computational power of continuous-time symmetric Hopfield nets. Since the dynamics of such networks are governed by Liapunov (energy) functions, they cannot generate infinite nondamping oscillations, and hence cannot simulate arbitrary (potentially divergent) discrete computations. Nevertheless, we prove that any convergent fully parallel computation by a network of $n$ discrete-time binary neurons, with in general asymmetric interconnections, can be simulated by a symmetric continuous-time Hopfield net containing $14n+6$ units using the saturated-linear sigmoid activation function. In terms of standard discrete computation models this result implies that any polynomially space-bounded Turing machine can be simulated by a polynomially size-increasing sequence of continuous-time Hopfield nets. Similar techniques as here yield corresponding results on the convergence time and computational power of discrete-time Hopfield nets. Comments are welcome, Best regards, -- Jirka Sima ****************************************************************** Mgr. Jiri Sima, CSc. Department of Theoretical Informatics Institute of Computer Science, Czech Academy of Sciences Pod vodarenskou vezi 2, 182 07 Prague 8, Czech Republic phone: (+4202) 66 05 30 30 e-mail: sima at uivt.cas.cz fax: (+4202) 85 85 789 www: http://www.uivt.cas.cz/~sima ****************************************************************** From saadd at helios.aston.ac.uk Fri Feb 19 05:06:00 1999 From: saadd at helios.aston.ac.uk (saadd@helios.aston.ac.uk) Date: Fri, 19 Feb 1999 10:06:00 +0000 (GMT) Subject: ON-LINE LEARNING IN NEURAL NETWORKS Message-ID: <3166.199902191006@sun.aston.ac.uk> The following book is available from Cambridge University Press; see http://www.cup.cam.ac.uk/Scripts/webbook.asp?isbn=0521652634 ON-LINE LEARNING IN NEURAL NETWORKS David Saad, Ed. Series: Publications of the Newton Institute ISBN: 0 521 65263 4 DESCRIPTION On-line learning is one of the most powerful and commonly used techniques for training large layered networks, and has been used successfully in many real-world applications. Traditional analytical methods have been recently complemented by ones from statistical physics and Bayesian statistics. This powerful combination of analytical methods provides more insight and deeper understanding of existing algorithms, and leads to novel and principled proposals for their improvement. This book presents a coherent picture of the state-of-the-art in the theoretical analysis of on-line learning. An introduction relates the subject to other developments in neural networks and explains the overall picture. Surveys by leading experts in the field combine new and established material and enable non-experts to learn more about the techniques and methods used. This book, the first in the area, provides a comprehensive view of the subject and will be welcomed by mathematicians, scientists and engineers, whether in industry or academia. CONTENTS Foreward - Christopher M. Bishop 1. Introduction David Saad 2. On-line Learning and Stochastic Approximations Leon Bottou 3. Exact and Perturbative Solutions for the Ensemble Dynamics Todd K. Leen 4. A Statistical Study on On-line Learning Noboru Murata 5. On-line Learning in Switching and Drifting Environments with Application to Blind Source Separation Klaus-Robert Mueller, Andreas Ziehe, Noboru Murata and Shun-ichi Amari 6. Parameter Adaptation in Stochastic Optimization Luis B. Almeida, Thibault Langlois, Jose D. Amaral and Alexander Plakhov 7. Optimal On-line Learning in Multilayer Neural Networks David Saad and Magnus Rattray 8. Universal Asymptotics in Committe Machines with Tree Architecture Mauro Copelli and Nestor Caticha 9. Incorporating Curvature Information into On-line Learning Magnus Rattray and David Saad 10. Annealed On-line Learning in Multilayer Neural Networks Siegfried Boes and Shun-ichi Amari 11. On-line Learning of Prototypes and Principal Components Michael Biehl, Ansgar Freking, Matthias Hoelzer, Georg Reents and Enno Schloesser 12. On-line Learning whith Time-Correlated Examples Tom Heskes and Wim Wiegerinck 13. On-line Learning from Finite Training Sets David Barber and Peter Sollich 14. Dynamics of Supervised Learning with Restricted Training Sets Anthony C.C. Coolen and David Saad 15. On-line Learning of a Decision Boundary with and without Queries Yoshiyuki Kabashima and Shigeru Shinomoto 16. A Bayesian Approach to On-line Learning Manfred Opper 17. Optimal perceptron learning: an on-line Bayesian approach Sara A. Solla and Ole Winther See: http://www.ncrg.aston.ac.uk/books/OLNN/index.html From simon.schultz at anu.edu.au Fri Feb 19 17:40:35 1999 From: simon.schultz at anu.edu.au (Simon R. Schultz) Date: Sat, 20 Feb 1999 09:40:35 +1100 Subject: Preprint available Message-ID: <36CDE863.8F5D029B@anu.edu.au> The following paper has been accepted for publication in Proceedings of the Royal Society Series B: Biological Sciences, and is available in preprint form: "Correlations and the encoding of information in the nervous system", S. Panzeri*, S. R. Schultz*, A. Treves+ and E. T. Rolls* * Department of Experimental Psychology, University of Oxford, South Parks Rd., Oxford OX1 3UD, U.K. + SISSA - Programme in Neuroscience, 34013 Trieste, Italy ------ Abstract: ------ Is the information transmitted by an ensemble of neurons determined solely by the number of spikes fired by each cell, or do correlations in the emission of action potentials also play a significant role? We derive a simple formula which enables this question to be answered rigorously for short timescales. The formula quantifies the corrections to the instantaneous information rate which result from correlations in spike emission between pairs of neurons. The mutual information the ensemble of neurons conveys about external stimuli can thus be broken down into firing rate and correlation components. This analysis provides fundamental constraints upon the nature of information coding -- showing that over short timescales, correlations cannot dominate information representation, that stimulus-independent correlations may lead to synergy (where the neurons together convey more information than they would considered independently), but that only certain combinations of the different sources of correlation result in significant synergy rather than in redundancy or in negligible effects. This analysis leads to a new quantification procedure which is directly applicable to simultaneous multiple neuron recordings. This paper may be downloaded from the web page http://www.cogneuro.ox.ac.uk/~schultz/corrinfo.html, in either compressed postscript or PDF format. -- Simon Schultz, D.Phil. Visiting Fellow, Psychobiology Laboratory, Division of Psychology, Building 38, Australian National University, Canberra, ACT 0200, Australia. Phone: +61-2 6279 9733 Fax: +61-2 6249 0499 http://www.cogneuro.ox.ac.uk/~schultz From S.Singh at exeter.ac.uk Fri Feb 19 13:10:22 1999 From: S.Singh at exeter.ac.uk (Sameer Singh) Date: Fri, 19 Feb 1999 18:10:22 +0000 (GMT Standard Time) Subject: Lectureship at Exeter University, UK In-Reply-To: Message-ID: School of Engineering and Computer Science LECTURER IN COMPUTER SCIENCE VACANCY (ref number 6084) Further Particulars We are seeking an active, energetic researcher to contribute to one of the research areas in Computer Science: artificial intelligence, distributed and parallel systems, media computing, pattern recognition and neural computing. We would expect the successful candidate to contribute to the teaching programme at all levels and to develop modules in their own area of interest. This is a permanent position, available from the Summer Term 1999, or as soon as possible thereafter. The School The School of Engineering and Computer Science was formed on 1 August 1998 through the merger of the former Department of Computer Science and School of Engineering. For the purposes of student recruitment and quality assessment, the subject departments of Computer Science and Engineering retain an identity within the new School. For management and other purposes, the School is a fully integrated academic unit under the Head of School, Professor David Owens. School-level committees handle both research and teaching policy matters. The philosophy of the new School is to promote interdisciplinary work across the boundaries of engineering and computing and between the engineering sub-disciplines. The approach has been in place within engineering since the formation (by merger of separate departments) of the former School of Engineering in 1988. The merger with the Department of Computer Science, founded in 1979, represents an exciting opportunity to extend this concept at all levels: undergraduate, postgraduate and staff. One of our objectives is to establish a degree programme that bridges the two disciplines: a major/minor degree in Electronics with Computer Science. At present third year students in Computer Science regularly take options within Engineering. The School also runs full MSc, MPhil and PhD programmes. At present, Computer Science and Engineering are located in separate buildings on the main University campus, but plans are being developed to bring the whole School into single premises with the Harrison building (the current home of Engineering). The relocation will provide the School, by the summer of 2000, with new, extensive, purpose-built laboratory and computing facilities. Computer Science There are currently eleven academic staff in Computer Science. All academic staff are expected to contribute to research and teaching and to undertake appropriate administrative duties. A Computer Support Group of seven, including a hardware technician, maintains and upgrades the Department's computing facilities in conjunction with the University's IT Services. Members of the group are able also to provide support to specific research projects. In our major undergraduate laboratories we run Windows 98/95 and Linux. In support of these clients, clusters of Silicon Graphics workstations act as servers connected by Samba, a public domain Internet implementation of a Windows NT service. The research laboratories generally use Unix, either PCs running Linux, Silicon Graphics running Irix or Sun workstations running SunOS or Solaris. In total, we run 16 Silicon Graphics workstations connected by an ATM network capable of high speed video transmission, 11 Sun workstations, 50 Pentium PCs including 27 multimedia PCs running Windows 98/95 or Linux, 4 laser printers including a high quality colour laser printer. Supported applications include software for mathematical computing, 3D animation and visualisation, parallel and distributed systems, image processing and neural networks. All our computers are connected to the Internet. We run our own web server including an internal Intranet. Research groups have their own specialist servers connected to the network. The University's IT Services provides undergraduate accounts and filespace and other specialist facilities. We maintain a World Wide Web entry with home page: http://www.dcs.exeter.ac.uk Details of the syllabus for the current taught modules may be found under the syllabus entry on the page: http://www.dcs.exeter.ac.uk/wyska Research Computer Science was graded 4A in the 1996 Research Assessment Exercise. While recognising the importance of fostering individual initiative in research, the school encourages the formation of groups which act as a stimulus to research, enable informed but informal critical discussion and provide a supportive environment for postgraduate study. We currently recognise three research groups in Computer Science: Pattern Recognition and Neural Computing (Director: Dr Sameer Singh) Pattern Analysis and Neural Networks represent technologies of the future at the cutting edge of computer science research. This group investigates research issues in a number of applied topics in computer science and engineering. Our interests extend to Pattern Recognition, Neural Networks, Image Processing, Medical Imaging, OCR, Financial Forecasting and Linguistic Computing. A large number of our projects are with external collaborators and are supported by both internal and external grants. The group is committed to excellence in research achieved through a range of interactions at national and international level. At present the research group has more than twelve full-time members. Full details about the group are available at the web site: http://www.dcs.ex.ac.uk/research/pann Distributed and Parallel Systems (Director: Dr Stephen Turner) Research within this group concentrates on programming systems formed by connecting a number of processors, where the connection could be as tight as the bus of a multiprocessor, or as loose as the Internet. Our work on parallel computing involves harnessing many processors to solve a single problem, with discrete event simulation being an area of particular interest. As well as running parallel programs, we are interested in predicting their performance on different machine configurations. Another area of interest is the successful exploitation of distributed database technology, where we research into its application in solving extant problems in complex information sharing domains. Finally, we are also interested in programming networked embedded computing devices using the idea of mobile code. Artificial Intelligence (Director: Dr Antony Galton) The work of this research group comprises three main strands:- Knowledge Representation, with particular application to qualitative spatial and temporal reasoning; Media Computing, including natural language engineering and information retrieval, GIS and visualisation, multimedia databases, and psychological aspects of multimedia; Future Computing Paradigms, such as evolutionary computing and genetic algorithms, and quantum computing In accordance with our research strategy, special purpose laboratories have been established with dedicated equipment within the first two of these sub-areas. We also maintain a Media Computing laboratory through which we explore a range of issues concerning the use of media in the way we communicate and undertake creative endeavours. Visualisation in particular - data, program, knowledge - features in all research groupings. We pride ourselves on presenting a seamless view of computational paradigms and treat all such paradigms (symbolic as well as sub-symbolic, parallel as well as sequential, declarative as well as procedural, object-oriented as well as block-structured) as equally valid, according to the circumstances. With the formation of the new School, we aim to enhance joint research initiatives. Initially these might be on the basis of engineering applications providing a demonstrator for software technologies developed in computer science. Also the theoretical backing provided by research in computer science could aid developments in engineering. There will also be possibilities for technology led developments in media computing. It is the responsibility of the Director of each research group to produce an annual research plan based on the individual plans of the staff within the group. Such plans are the basis for recommendations on equipment purchase and access to travel funds and study leave. There are currently sixteen students on our MPhil/PhD programme. Since 1989 we have had a 100% submission record for all SERC/EPSRC funded PhD students. We run a thesis committee system whereby each student is encouraged in the development of their research but also has their progress monitored. We have an EPSRC-approved advanced, specialist MSc degree programme in New Generation Computing. Undergraduate Teaching We received a rating of excellent in 1994 HEFCE Teaching Quality Assessment, one of only 10 UK Universities so rated. Our submission to HEFCE stressed the importance of feeding research back into the teaching programmes and we apply this approach to all our degrees. Approximately 65 undergraduates are admitted each year. Our main programme is the Single Honours degree in Computer Science. In collaboration with the School of Psychology we offer an Interdisciplinary Single Honours programme in Cognitive Science. There are also in place arrangements for major/minor programmes with the School of Mathematical Sciences and a with European study programme. We have placed some of our modules within the University's modularity scheme thus allowing students from other disciplines to study computing. Teaching allocations are based on a formula that takes account of research supervisory commitments. Also, all junior staff are given time to undertake a structured independent-learning course (recognised by the Staff and Educational Development Association (SEDA)) run by the Staff Development Unit. Other Information The starting salary will be up to £26,523 on the in the Lecturer A scale or Lecturer B scale (£16,655 - £29,048), dependent upon teaching research experience. There is a 3-year probationary period. Full-time members of the academic staff may not undertake without consent of the University Council outside paid work, except external examining, literary work, public lecturing, and extra-mural teaching, occasional broadcasting and the giving of advice on an ad-hoc basis and on a relatively small scale. The successful applicant (if aged under 60) may join the Universities' Superannuation Scheme (USS), or maintain USS membership if applicable; 6.35% of gross salary is contributable by the individual, with the University contributing such sums as will be required to maintain the full benefits of the scheme. For permanent appointments, in the case of a person appointed from within the United Kingdom, all agreed removal expenses will normally be reimbursed. In the case of the appointment of a candidate from overseas, the full cost of removal expenses together with one adult economy fare from the candidate's place of residence will normally be reimbursed. In this context, it should be noted that the phrase "removal expenses" is interpreted by the University to mean the removal of household effects and excludes the cost of storage and the reimbursement of personal fares save the one adult economy fare payable to a person appointed overseas. Claims for reimbursement of removal expenses should be accompanied by three competitive tenders and the receipted account. Reimbursement will normally be on the basis of the lowest tender. Interviewees will be asked to complete a Medical Questionnaire. The University is committed to a comprehensive policy of equal opportunities in employment in which individuals are selected and treated solely on the basis of their relevant merits and abilities and are given equal opportunities within the University. In making this appointment the University reserves the right to consider persons other than those who have submitted applications by the closing date. To apply, candidates forward 7 copies (or, if posted overseas, 1 copy in a format suitable for photocopying) of your curriculum vitae, including particulars of your qualifications and experience, and a list of their publications, together with names and addresses of three referees a supporting letter of application which includes a statement indicating your intended areas of research, and summary of personal details sheet. Applications should reach the Personnel Division, University of Exeter, Northcote House, The Queen's Drive, Exeter, EX4 4QJ, Devon, UK, by 26 February 1999 . Please quote the reference number 6084 on your application and in all correspondence. Candidates should note that the University normally contacts referees for shortlisted candidates without further advice to applicants. When advising of referees, if they are resident overseas please ensure you supply fax numbers and/or email addresses enabling us to contact them (if necessary) without delay. In the interests of economy, the University does not notify candidates who have been unsuccessful with their application; consequently if you have not heard from us within six weeks of the closing date, please assume that you have not been successful on this occasion. For further information and an informal discussion please contact Dr Wendy Milne (w.milne at exeter.ac.uk), telephone 01392 264061. Informal contact with the Directors of research is also welcomed. fps.6084 From l.s.smith at cs.stir.ac.uk Fri Feb 19 08:28:06 1999 From: l.s.smith at cs.stir.ac.uk (Dr L S Smith (Staff)) Date: Fri, 19 Feb 99 13:28:06 GMT Subject: deadline extended for ICANN 99 - Ninth International Conference on Artificial Neural Networks Message-ID: <199902191328.NAA20500@tinker.cs.stir.ac.uk> ICANN 99 Submission Deadline - Extended ICANN 99, the annual conference of the European Neural Network Society held in conjunction with the IEE conference on Artificial Neural Networks, is taking place in Edinburgh, UK, from 07-10 September 1999. Due to numerous requests for bringing the ICANN 99 submission procedure into alignment with earlier years, the deadline for submissions has been revised to Friday 12 March 1999. For more information see http://www.iee.org.uk/Conf/ICANN/Welcome.html and for post-conference workshops see http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/icann/ Leslie Smith Alan Murray David Willshaw For the Organising Committee From harnad at coglit.soton.ac.uk Sun Feb 21 08:04:58 1999 From: harnad at coglit.soton.ac.uk (Stevan Harnad) Date: Sun, 21 Feb 1999 13:04:58 +0000 (GMT) Subject: Law of Effect: BBS Call for Commentators Message-ID: Below is the abstract of a forthcoming BBS target article *** please see also 5 important announcements about new BBS policies and address change at the bottom of this message) *** BEHAVIORAL MOMENTUM AND THE LAW OF EFFECT by John A. Nevin & Randolph C. Grace This article has been accepted for publication in Behavioral and Brain Sciences (BBS), an international, interdisciplinary journal providing Open Peer Commentary on important and controversial current research in the biobehavioral and cognitive sciences. Commentators must be BBS Associates or nominated by a BBS Associate. To be considered as a commentator for this article, to suggest other appropriate commentators, or for information about how to become a BBS Associate, please send EMAIL before March 26 to: bbs at cogsci.soton.ac.uk or write to [PLEASE NOTE SLIGHTLY CHANGED ADDRESS]: Behavioral and Brain Sciences ECS: New Zepler Building University of Southampton Highfield, Southampton SO17 1BJ UNITED KINGDOM http://www.princeton.edu/~harnad/bbs/ http://www.cogsci.soton.ac.uk/bbs/ ftp://ftp.princeton.edu/pub/harnad/BBS/ ftp://ftp.cogsci.soton.ac.uk/pub/bbs/ If you are not a BBS Associate, please send your CV and the name of a BBS Associate (there are currently over 10,000 worldwide) who is familiar with your work. All past BBS authors, referees and commentators are eligible to become BBS Associates. To help us put together a balanced list of commentators, please give some indication of the aspects of the topic on which you would bring your areas of expertise to bear if you were selected as a commentator. An electronic draft of the full text is available for inspection with a WWW browser, anonymous ftp or gopher according to the instructions that follow after the abstract. _____________________________________________________________ BEHAVIORAL MOMENTUM AND THE LAW OF EFFECT John A. Nevin University of New Hampshire RR2, Box 162, Vineyard Haven, MA 02568 USA tnevin at worldnet.att.net Randolph C. Grace, University of Canterbury Department of Psychology, Christchurch, New Zealand r.grace at psyc.canterbury.ac.nz KEYWORDS: addiction, choice, extinction, generalization, law of effects, learning, momentum, movement, operant, Pavlov, preference, reinforcement, self-control, Thorndike ABSTRACT: In the metaphor of behavioral momentum, the rate of a free operant in the presence of a discriminative stimulus is analogous to the velocity of a moving body, and resistance to change measures an aspect of behavior that is analogous to its inertial mass. An extension of the metaphor suggests that preference measures an analog to the gravitational mass of that body. The independent functions relating resistance to change and preference to the conditions of reinforcement may be construed as convergent measures of a single construct, analogous to physical mass, that represents the effects of a history of exposure to the signaled conditions of reinforcement and that unifies the traditionally separate notions of the strength of learning and the value of incentives. Research guided by the momentum metaphor emcompasses the effects of reinforcement on response rate, resistance to change, and preference, and has implications for clinical interventions, drug addiction, and self-control. In addition, its principles can be seen as a modern, quantitative version of Thorndike's (1911) Law of Effect, providing a new perspective on some of the challenges to his postulation of strengthening by reinforcement. ____________________________________________________________ To help you decide whether you would be an appropriate commentator for this article, an electronic draft is retrievable from the World Wide Web or by anonymous ftp from the US or UK BBS Archive. Ftp instructions follow below. Please do not prepare a commentary on this draft. Just let us know, after having inspected it, what relevant expertise you feel you would bring to bear on what aspect of the article. The URLs you can use to get to the BBS Archive: http://www.princeton.edu/~harnad/bbs/ http://www.cogsci.soton.ac.uk/bbs/Archive/bbs.nevin.html ftp://ftp.princeton.edu/pub/harnad/BBS/bbs.nevin ftp://ftp.cogsci.soton.ac.uk/pub/bbs/Archive/bbs.nevin To retrieve a file by ftp from an Internet site, type either: ftp ftp.princeton.edu or ftp 128.112.128.1 When you are asked for your login, type: anonymous Enter password as queried (your password is your actual userid: yourlogin at yourhost.whatever.whatever - be sure to include the "@") cd /pub/harnad/BBS To show the available files, type: ls Next, retrieve the file you want with (for example): get bbs.nevin When you have the file(s) you want, type: quit ____________________________________________________________ *** FIVE IMPORTANT ANNOUNCEMENTS *** ------------------------------------------------------------------ (1) There have been some very important developments in the area of Web archiving of scientific papers very recently. Please see: Science: http://www.cogsci.soton.ac.uk/~harnad/science.html Nature: http://www.cogsci.soton.ac.uk/~harnad/nature.html American Scientist: http://www.cogsci.soton.ac.uk/~harnad/amlet.html Chronicle of Higher Education: http://www.chronicle.com/free/v45/i04/04a02901.htm --------------------------------------------------------------------- (2) All authors in the biobehavioral and cognitive sciences are strongly encouraged to archive all their papers (on their Home-Servers as well as) on CogPrints: http://cogprints.soton.ac.uk/ It is extremely simple to do so and will make all of our papers available to all of us everywhere at no cost to anyone. [Note, this is not addressed particularly to BBS authors, but to ALL authors of ALL articles in the biobehavioral and cognitive and related sciences.] --------------------------------------------------------------------- (3) BBS has a new policy of accepting submissions electronically. Authors can specify whether they would like their submissions archived publicly during refereeing in the BBS under-refereeing Archive, or in a referees-only, non-public archive. Upon acceptance, preprints of final drafts are moved to the public BBS Archive: ftp://ftp.princeton.edu/pub/harnad/BBS/.WWW/index.html http://www.cogsci.soton.ac.uk/bbs/Archive/ -------------------------------------------------------------------- (4) BBS has expanded its annual page quota and is now appearing bimonthly, so the service of Open Peer Commentary can now be be offered to more target articles. The BBS refereeing procedure is also going to be considerably faster with the new electronic submission and processing procedures. Authors are invited to submit papers to: Email: bbs at cogsci.soton.ac.uk Web: http://cogprints.soton.ac.uk http://bbs.cogsci.soton.ac.uk/ INSTRUCTIONS FOR AUTHORS: http://www.princeton.edu/~harnad/bbs/instructions.for.authors.html http://www.cogsci.soton.ac.uk/bbs/instructions.for.authors.html --------------------------------------------------------------------- (5) Call for Book Nominations for BBS Multiple Book Review In the past, Behavioral and Brain Sciences (BBS) journal had only been able to do 1-2 BBS multiple book treatments per year, because of our limited annual page quota. BBS's new expanded page quota will make it possible for us to increase the number of books we treat per year, so this is an excellent time for BBS Associates and biobehavioral/cognitive scientists in general to nominate books you would like to see accorded BBS multiple book review. (Authors may self-nominate, but books can only be selected on the basis of multiple nominations.) It would be very helpful if you indicated in what way a BBS Multiple Book Review of the book(s) you nominate would be useful to the field (and of course a rich list of potential reviewers would be the best evidence of its potential impact!). From mw at stat.Duke.EDU Mon Feb 22 11:18:26 1999 From: mw at stat.Duke.EDU (Mike West) Date: Mon, 22 Feb 1999 11:18:26 -0500 Subject: 1999 MITCHELL PRIZE: revisited Message-ID: <19990222111826.A14866@isds.duke.edu> This is a reposting of the 1999 Mitchell Prize announcement, correcting a typo and clarifying eligibility. With apologies for multiple postings, we'd appreciate your help in bringing this to the attention of potential nominators and future Mitchell Prize winners! Thanks for your help. ---------------------------------------------------------------------- 1999 MITCHELL PRIZE: ANNOUNCEMENT AND SOLICITATION http://www.stat.duke.edu/sites/mitchell.html The third Mitchell Prize will be awarded in 1999. The Prize will be announced and presented at the August 1999 Joint Statistical Meetings in Baltimore. The Prize is awarded in recognition of an outstanding paper describing how a Bayesian analysis has solved an important applied problem. The 1999 Prize consists of an award of $1000 and a commemorative plaque. The Mitchell Prize is named for Toby J. Mitchell and was established by his friends and colleagues following his death from leukemia in 1993. Toby was a Senior Research Staff Member at Oak Ridge National Laboratory throughout his career, will leaves of absence spent at the University of Wisconsin and at the National Institute of Environmental Health Sciences. Toby won the Snedecor Award in 1978 (with co-author Bruce Turnbull), made incisive contributions to statistics, especially in biometry and engineering applications, and was a marvelous collaborator and an especially thoughtful scientist. Toby was a dedicated Bayesian, hence the focus of the prize. This is the third Mitchell Prize, the first two having been awarded in 1994 and 1997. Beginning this year, the Prize will be awarded annually under the cosponsorship of the ASA Section on Bayesian Statistical Science (SBSS), the International Society for Bayesian Analysis (ISBA), and the Mitchell Prize Founders' Committee. The sponsors annually establish a selection committee; the 1999 Prize selection committee consists of Gary Koop, Max Morris (chair) and Mike West. To be eligible for the 1999 Prize, a paper will either have appeared in a refereed journal or refereed conference proceedings since January 1 1997, or be scheduled for future publication in a refereed outlet. Candidate papers will be accepted from nominators and from authors. In reviewing submissions, emphasis will be placed on evidence that the application has truly benefited from a Bayesian analysis respecting the individual character of the problem at hand. There is no restriction as to approach taken, except that it be Bayesian in some sense, and that it carefully and appropriately justifies models, priors and methodologies adopted. To be considered for this year's Mitchell Prize, please send four reprints or copies of the manuscript along with a cover letter that includes: a brief statement of the impact of the work, and contact information for the authors, nominator (if not an author), and two individuals who are not statisticians or coauthors and who can be contacted for an evaluation of the importance of the work in the applied field. Submissions should be mailed to Max Morris Department of Statistics Iowa State University Snedecor Hall Ames, IA 50011-1210 Entries must be received by May 15, 1999 in order to receive consideration. For further information, phone 515-294-2775, FAX 515-294-4040, or e-mail mmorris at iastate.edu. To learn more about the sponsoring organizations SBSS and ISBA visit their web sites at www.stat.duke.edu/sbss/sbss.html and www.bayesian.org/ respectively. ------------------------------------------------------------------------- From istvan at usl.edu Mon Feb 22 17:48:47 1999 From: istvan at usl.edu (Dr. Istvn S. N. Berkeley) Date: Mon, 22 Feb 1999 16:48:47 -0600 Subject: Cog. Sci. Job Message-ID: <36D1DECF.37D@USL.edu> The University of Southwestern Louisiana's Institute for Cognitive Science invites applications for a tenure-track faculty appointment at the assistant professor level for a new Ph.D. program. Appointment is anticipated for the Fall of 1999. Applicants must have an earned doctorate in cognitive science or a closely related area, as well as demonstrated potential for research excellence. Focus areas of the new program are in cognitive processes, comparative cognition, computational models of mind, cognitive neuroscience, and language and thought. Salary is competitive, depending on experience and qualifications. Applicants should send a letter of application, statement of research interests, curriculum vitae, reprints, and three letters of reference to Dr. Daniel Povinelli, Institute for Cognitive Science, P.O. Drawer 43772, University of Southwestern Louisiana, Lafayette, Louisiana 70504. Review of applications will commence March 15, 1999. The University is in compliance with Title IX of the Civil Rights Act, Section 504 of the Rehabilitation Act of 1973, and is an Equal Employment Opportunity Affirmative Action Employer. -- Istvan S. N. Berkeley Ph.D, E-mail: istvan at USL.edu, Philosophy, The University of Southwestern Louisiana, USL P. O. Box 43770, Lafayette, LA 70504-3770, USA. Tel:(318) 482 6807, Fax: (318) 482 6195, http://www.ucs.usl.edu/~isb9112 From amari at brain.riken.go.jp Tue Feb 23 05:27:25 1999 From: amari at brain.riken.go.jp (Shunichi Amari) Date: Tue, 23 Feb 1999 19:27:25 +0900 Subject: Brain Science Summer Course Message-ID: <19990223192725G.amari@brain.riken.go.jp> The RIKEN Brain Science Institute is offering a summer program to train advanced students interested in molecular, cellular, systems, or computational neuroscience. Applicants may choose either a laboratory internshipfor up to two months, or participate in an intensive two-week lecture course featuring a distinguished international faculty. See our website for details. Summer Internship: July - September, 1999 Lecture Course: July 25 - August 7, 1999 http://www.brain.riken.go.jp/summer/bsi-summer.html Shun-ichi Amari From jose.m.bernardo at uv.es Tue Feb 23 06:41:16 1999 From: jose.m.bernardo at uv.es (Jose M. Bernardo) Date: Tue, 23 Feb 1999 12:41:16 +0100 Subject: Workshop of Bayesian Objective Methodology Message-ID: Apologies for cross-posting INTERNATIONAL WORKSHOP ON OBJECTIVE BAYESIAN METHODOLOGY SECOND INFORMATION BULLETIN 1499--1999: Five Centuries of the Universitat de Val?ncia Valencia, Spain, June 11th -- June 13th, 1999 The Universitat of Val?ncia, Spain, will celebrate next year its 500 anniversary. Within the Centenary Programme, the University of Valencia is sponsoring an International Workshop on Objective Bayesian Methodology Objective Bayesian Methodology: Objective Bayesian methodology is of increasing importance today for at least two reasons. First, application of Bayesian analysis is rapidly growing among nonspecialists, most of whom seek automatic or objective Bayesian procedures. Second, computational advances have allowed Bayesian methodology to be employed in problems of such complexity that determination of serious subjective priors is essentially impossible. The corresponding explosion of research on objective Bayesian methods makes a workshop in this area particularly timely. Objective Bayesian methodology is, for the most part, oriented towards the development of prior distributions that can be used automatically, ie that do not require subjective input. There are three quite distinct statistical domains in which this development has taken place: parametric estimation, parametric testing, and nonparametric analysis. The specific topics in these areas that are currently being considered for the Workshop are listed below. Note, however, that it is planned to maintain considerable flexibility so as to be able to include late-breaking developments. 1. Objective Priors and Frequentist Statistics 2. Determination of Objective Priors in Important Problems 3. Priors for Objective Bayesian Testing and Model Selection 4. Objective Priors in Nonparametric Analysis 5. The Roles of Objective Bayesian Analysis Workshop Organization: The three day workshop will take place at the Peset Center, a reconverted XVIII century palace in the medieval old town of the city of Valencia, with appropriate facilities. Workshop Web sites: http://www.uv.es/~bernardo/workshop.html (Spain) http://www.stat.duke.edu/~bernardo/workshop.html (USA mirror) Programme Committee: Professor James O. Berger ({ Duke University, USA}) Professor Jos? M. Bernardo ({ Universitat de Val?ncia, Spain}) Professor Jayanta K. Ghosh ({ Indian Statistical Institute, India and Purdue University, USA}) Professor Malay Ghosh ({ University of Florida, USA}) Professor Christian P. Robert ({ Universit? de Rouen, France}) Scientific Programme: The scientific programme will feature the following invited lectures Berger, James, Duke University, USA (Generalized) Reference Priors for Spatial Models (with B. Sanso, and V. De Oliveira) Bernardo, Jos?, Universidad de Valencia, Spain Bayesian Reference Multivariate Hypothesis Testing (with Rueda) Datta, Gauri, University of Georgia, USA Bayesian Prediction with Approximate Frequentist Validity Dawid, a. Phil, University College London, UK The Trouble with Bayes Factors Fraser, Donald, University of Toronto, Canada Objective priors and model dependence: the frequentist-Bayesian interface Ghosh, Jayanta, Indian Statistical Institute, India Prior Elicitation, Default Priors and Prediction Loss Ghosh, Malay, University of Florida, USA Noninformative Priors for Item Response Models Ghosal, Subhashis, Vrije Universiteit Amsterdam, the Netherlands Default Priors, Model Selection and Adaptation in Density Estimation Liseo, Brunero, Universit? di Roma La Sapienza, Italy, Semiparametric Inference for Long Memory Time Series: on Bayesian-Frequentist Interface (with D. Marinucci and L. Petrella) Mukerjee, Rahul, Indian Institute of Management, India On Confidence Intervals Associated with the Usual and Adjusted Likelihoods (with N. Reid) P?rez, Jos? M., Universidad Simon Bolivar, Venezuela Expected Posterior Priors for Default Model Selection Pericchi, Luis, Universidad Sim?n Bol?var, Venezuela The Intrinsic Bayes Factor as a Constructive Method to Develop Default or Conventional Priors for Hypothesis Testing and Model Selection. Robert, Christian, CREST-ENSAE, France Nonidentifiable Models and Symmetric Priors: How to Teach MCMC Samplers to Jump Farther (with M. Hurn) Rousseau, Judith, CREST-ENSAE, France Edgeworth expansion for a Gaussian long-memory process with application to matching priors Sun, Dongchu, University of Missouri-Columbia, USA Random Effects in Generalized Linear Mixed Models Sweeting, Trevor, University of Surrey, UK Coverage Probability Bias, Objective Bayes and the Likelihood Principle Wasserman, Larry, Carnegie Mellon University, USA Priors for Infinite Dimensional Problems Each lecture will be followed by invited discussions and by discussion from the floor. Invited discussants include: Barbieri, Marilena, Universit? di Roma La Sapienza, Italy Bayarri, Susie, Universitat de Val?ncia, Spain Clarke, Bertrand, University of British Columbia, Canada Clyde, Merlise, Duke University , USA de Santis, Fulvio, Universit? di Roma La Sapienza, Italy DiCiccio, Thomas, Stanford University, USA Gelfand, Alan, University of Connecticut, USA George, Edward, University of Texas, USA Gir?n, Javier, Universidad de M?laga, Spain Guti?rrez-Pe?a, Eduardo, UNAM, Mexico Mendoza, Manuel, ITAM, Mexico Moreno, El?as, Universidad de Granada, Spain Mortera, Julia, Universit? di Roma 3, Italy Petrone, Sonia, Universit? di Pavia, Italy Ramamoorthi, R., Michigan State University, USA Reid, Nancy, University of Toronto, Canada Rueda, Ra?l, UNAM, Mexico Severini, Thomas, Northwestern University, USA Sivaganesan, Siva, University of Cincinnati, USA van der Linde, Angelika, University of Edinburgh, UK Wells, Martin, Cornell University, USA Ye, Keying, Virginia Polytechnic Institute and State University, USA Contributed papers are encouraged. These will be presented in two plenary poster sessions, organized in the tradition of the Valencia meetings. The abstracts deadline is April 30th . The abstract should be sent by e-mail using the abstract format http://www.uv.es/~bernardo/abstract.html (Spain) http://www.stat.duke.edu/~bernardo/abstract.html (USA mirror) accessible at the workshop web sites. There will not be published proceedings, but a booklet of extended abstracts will be distributed at the Workshop. Travel Information: Delegates will be expected to arrive on Thursday June 10th evening, and to depart on Monday 14th morning. The opening ceremony will be on the morning of Friday 11th, and the Conference Dinner on the evening of Sunday 13th. Valencia international airport, 7 km. from town, has direct connections to many european cities, and frequent connexions to both Madrid and Barcelona. Registration: To register, please mail or fax the registration form http://www.uv.es/~bernardo/registration.html (Spain) http://www.stat.duke.edu/~bernardo/registration.html (USA mirror) accessible at the workshop web sites. Those possibly interested in participating, but not ready to register, should e-mail the local organizer, Jos? M. Bernardo mailto:jose.m.bernardo at uv.es Periodic electronic Information Bulletins, will be sent to those who who have declared their interest. Additional information: Further information on the workshop, including pointers to the first draft of the talks as they become available is posted in the workshop web sites. http://www.uv.es/~bernardo/workshop.html (Spain) http://www.stat.duke.edu/~bernardo/workshop.html (USA mirror) For tourist information on the area, please visit he Web site of the Tourism Agency of the State of Valencia, http://www.comunitat-valenciana.com PLEASE NOTE: On April 4th, all Spanish telephone numbers added a "9" after the country code "34'. Thus, my more direct (phone and fax) line is now: +34.96.364.3560 ____________________________________________________________________ Prof. Dr. Jose M. Bernardo, Universidad de Valencia Dept. Estadistica, Fac. Matematicas, 46100-Burjassot, Valencia, Spain Tel. +34.96.364.3560 and +34.96.386.4313 (direct), +34.96.386.4362 (office). Fax: +34.96.364.3560 (direct), +34.96.386.4735 (office). E Mail: jose.m.bernardo at uv.es, Web Pages: http://www.uv.es/~bernardo/ or http://www.stat.duke.edu/~bernardo/ ____________________________________________________________________ From Jerry.Friedman at cmis.CSIRO.AU Tue Feb 23 21:12:34 1999 From: Jerry.Friedman at cmis.CSIRO.AU (Jerry.Friedman@cmis.CSIRO.AU) Date: Wed, 24 Feb 1999 13:12:34 +1100 Subject: Boosting methods for regression and classification. Message-ID: <199902240212.NAA18316@pride.nsw.cmis.CSIRO.AU> *** Technical Report Available *** Greedy Function Approximation: A Gradient Boosting Machine Jerome H. Friedman Stanford University ABSTRACT Function approximation is viewed from the perspective of numerical optimization in function space, rather than parameter space. A connection is made between stagewise additive expansions and steepest-descent minimization. A general gradient-descent "boosting" paradigm is developed for additive expansions based on any fitting criterion. Specific algorithms are presented for least-squares, least-absolute-deviation, and Huber-M loss functions for regression, and multi-class logistic likelihood for classification. Special enhancements are derived for the particular case where the individual additive components are decision trees, and tools for interpreting such "TreeBoost" models are presented. Gradient boosting of decision trees produces competitive, highly robust, interpretable procedures for regression and classification, especially appropriate for mining less than clean data. Connections between this approach and the boosting methods of Freund and Shapire 1996, and Friedman, Hastie, and Tibshirani 1998 are discussed. Available from: "http://www-stat.stanford.edu/~jhf/ftp/trebst.ps" From Jon.Baxter at syseng.anu.edu.au Wed Feb 24 08:14:50 1999 From: Jon.Baxter at syseng.anu.edu.au (Jonathan Baxter) Date: Thu, 25 Feb 1999 00:14:50 +1100 (EST) Subject: Boosting methods for regression and classification. In-Reply-To: <199902240212.NAA18316@pride.nsw.cmis.CSIRO.AU> from "Jerry.Friedman@cmis.CSIRO.AU" at Feb 24, 99 01:12:34 pm Message-ID: <199902241314.AAA20350@reid.anu.edu.au> > > > > *** Technical Report Available *** > > > Greedy Function Approximation: > A Gradient Boosting Machine > > Jerome H. Friedman > Stanford University > > ABSTRACT > > Function approximation is viewed from the perspective of numerical > optimization in function space, rather than parameter space. A > connection is made between stagewise additive expansions and > steepest-descent minimization. A general gradient-descent "boosting" > paradigm is developed for additive expansions based on any fitting > criterion. > Specific algorithms are presented for least-squares, > least-absolute-deviation, and Huber-M loss functions for regression, > and multi-class logistic likelihood for classification. Special > enhancements are derived for the particular case where the individual > additive components are decision trees, and tools for interpreting > such "TreeBoost" models are presented. Gradient boosting of decision > trees produces competitive, highly robust, interpretable procedures > for regression and classification, especially appropriate for mining > less than clean data. Connections between this approach and the > boosting methods of Freund and Shapire 1996, and Friedman, Hastie, and > Tibshirani 1998 are discussed. > > > Available from: "http://www-stat.stanford.edu/~jhf/ftp/trebst.ps" There was also some discussion of the connection between boosting and gradient descent in function space at the NIPS workshop on large margins in December. I have put the slides of my talk on the subject---"AnyBoost: Boosting with (almost) arbitrary cost functions and steps"---on my web page for those who are interested (http://syseng.anu.edu.au/~jon/anyboost.ps). Cheers, Jon ------------- Jonathan Baxter Research Fellow Department of Systems Engineering Research School of Information Science and Engineering Australian National University http://syseng.anu.edu.au/~jon Tel: +61 2 6279 8678 Fax: +61 2 6279 8688 > > > From rsun at research.nj.nec.com Wed Feb 24 13:43:40 1999 From: rsun at research.nj.nec.com (Ron Sun) Date: Wed, 24 Feb 1999 13:43:40 -0500 Subject: TR on reinforcement learning Message-ID: <199902241843.NAA00660@pc-rsun.nj.nec.com> Announcing three new papers on combining reinforcement learning with symbolic methods: --------------------------------- Autonomous Learning of Sequential Tasks: Experiments and Analyses by Ron Sun$^{1,2}$, Todd Peterson$^2$ $^1$ NEC Research Institute, 4 Independence Way, Princeton, NJ 08540 \\ $^2$ The University of Alabama, Appeared in IEEE Transactions on Neural Networks, Nov.1998 http://cs.ua.edu/~rsun/sun.tnn98.ps ABSTRACT: This paper presents a novel learning model {\sc Clarion}, which is a hybrid model based on the two-level approach proposed in Sun (1995). The model integrates neural, reinforcement, and symbolic learning methods to perform on-line, bottom-up learning (i.e., learning that goes from neural to symbolic representations). The model utilizes both procedural and declarative knowledge (in neural and symbolic representations respectively), tapping into the synergy of the two types of processes. It was applied to deal with sequential decision tasks. Experiments and analyses in various ways are reported that shed light on the advantages of the model. --------------------------------- Multi-Agent Reinforcement Learning: Weighting and Partitioning by Ron Sun and Todd Peterson To appear in: Neural Networks http://cs.ua.edu/~rsun/sun.NN99.ps ABSTRACT: This paper addresses weighting and partitioning in complex reinforcement learning tasks, with the aim of facilitating learning. The paper presents some ideas regarding weighting of multiple agents and extends them into partitioning an input/state space into multiple regions with differential weighting in these regions, to exploit the differential characteristics of regions and the differential characteristics of agents to reduce the learning complexity of agents (and their function approximators) and thus to facilitate the learning overall. It analyzes, in reinforcement learning tasks, different ways of partitioning a task and using agents selectively based on partitioning. Based on the analysis, some heuristic methods are described and experimentally tested. We find that some off-line heuristic methods performed the best, significantly better than single-agent models. --------------------------------- A Hybrid Architecture for Situated Learning of Reactive Sequential Decision Making by Ron Sun Todd Peterson Edward Merrill To appear in: Applied Intelligence http://cs.ua.edu/~rsun/sun.apin99.ps ABSTRACT: In developing autonomous agents, one usually emphasizes only (situated) procedural knowledge, ignoring more explicit declarative knowledge. On the other hand, in developing symbolic reasoning models, one usually emphasizes only declarative knowledge, ignoring procedural knowledge. In contrast, we have developed a learning model {\sc Clarion}, which is a hybrid connectionist model consisting of both localist and distributed representations, based on the two-level approach proposed in Sun (1995). {\sc Clarion} learns and utilizes both procedural and declarative knowledge, tapping into the synergy of the two types of processes, and enables an agent to learn in situated contexts and generalize resulting knowledge to different scenarios. It unifies connectionist, reinforcement, and symbolic learning in a synergistic way, to perform on-line, bottom-up learning. This summary paper presents one version of the architecture and some results of the experiments. ----------------------------------------- Dr. Ron Sun NEC Research Institute 4 Independence Way Princeton, NJ 08540 phone: 609-520-1550 fax: 609-951-2483 email: rsun at cs.ua.edu, rsun at research.nj.nec.com (July 1st, 1998 -- July 1st, 1999) ----------------------------------------- Prof. Ron Sun http://cs.ua.edu/~rsun Department of Computer Science and Department of Psychology phone: (205) 348-6363 The University of Alabama fax: (205) 348-0219 Tuscaloosa, AL 35487 email: rsun at cs.ua.edu From B.Graham at anc.ed.ac.uk Thu Feb 25 12:59:03 1999 From: B.Graham at anc.ed.ac.uk (Bruce Graham) Date: Thu, 25 Feb 1999 17:59:03 GMT Subject: ICANN*99 Call for Tutorial Proposals Message-ID: <199902251759.RAA01257@dingo.cns.ed.ac.uk> CALL FOR PROPOSALS ICANN*99 Conference Tutorials 7 September 1999 Edinburgh, Scotland Tutorials on current topics in the fields of research covered by ICANN99 will be held on Tuesday 7 September 1999, the first day of the ICANN conference program. Proposals by qualified individuals interested in giving a half-day tutorial are solicited. Topics should fall within the broadly defined area of neural networks research, including: Theory and Algorithms; Neurobiology and Computational Neuroscience; Cognitive Modelling; Industrial, Commercial and Medical Applications; Hardware and Neuromorphic Engineering; Control, Robotics and Adaptive Behaviour. It is planned to run at least 6 tutorials of around 3 hours each: 3 in parallel in the morning and another 3 in the afternoon. Attendees will pay a registration fee and tutors will be given some recompense for their time and effort. Consequently we seek very high quality proposals. This call for proposals can be found at: http://www.iee.org.uk/Conf/ICANN/ and further information will be added as it becomes available. Submission Instructions ----------------------- Interested parties should submit via e-mail a short proposal for a tutorial of interest by 29 March, 1999 to Dr Bruce Graham (B.Graham at ed.ac.uk), Institute for Adaptive and Neural Computation, University of Edinburgh. Proposals should include an outline of the tutorial contents and your background and qualifications for giving such a tutorial. The final programme of tutorials will be decided by the middle of April. If e-mail is unavailable, mail or fax so as to arrive by the deadline to: ICANN*99 Tutorials c/o Dr Bruce Graham Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 2 Buccleuch Place, Edinburgh EH8 9LW, Scotland, UK Fax: +44-131-650-4406 PROPOSALS MUST BE RECEIVED BY MARCH 29 1999 (Bruce Graham, IANC, Division of Informatics, University of Edinburgh) From torras at iri.upc.es Thu Feb 25 07:29:43 1999 From: torras at iri.upc.es (Carme Torras) Date: Thu, 25 Feb 1999 13:29:43 +0100 (MET) Subject: Reminder: CFP Adaptive Robots Special Issue Message-ID: <199902251229.NAA05286@sibelius.upc.es> This is being mailed to several mailing lists. Please accept our apologies if you receive multiple copies. =============================================================================== Special Issue of the journal CONNECTION SCIENCE on *** ADAPTIVE ROBOTS *** CALL FOR PAPERS: Deadline March 15th, 1999 Adaptivity is the capability of self-modification that some agents have, which allows them to maintain a level of performance when facing environmental changes, or to improve it when confronted repeatedly with the same situation. This special issue is aimed at capturing the state of the art in the intricate task of endowing robots with adaptive capabilities, with a special emphasis on neural-based solutions. Thus, some examples of topics covered are: - Adaptive sensing - Adaptive gaits for walking robots - Self-calibration of robot manipulators - Adaptive dynamic control of flexible robot arms - Acquiring fine manipulation skills - Learning hand-eye coordination - Exploration and reinforcement learning - Improving robot navigation - Adaptive multi-robot systems The special issue will adhere to an engineering perspective, i.e. the emphasis will be on solving practical robotic problems using adaptive techniques, disregarding their possible biological (or cognitive) inspiration or plausibility. Work on real robots is preferred, with special attention being devoted to replicability of results, as well as to the discussion of the limitations (together with the advantages) of the proposed techniques. Guest editor: ------------- Carme TORRAS, CSIC-UPC (Spain) Editorial board: ---------------- Rudiger DILLMANN, University of Karlsruhe (Germany) Leslie P. KAELBLING, Brown University (USA) Ben KROSE, University of Amsterdam (The Netherlands) Jose R. MILLAN, Joint Research Centre (Italy) Helge RITTER, University of Bielefeld (Germany) Shankar SASTRY, University of California at Berkeley (USA) Noel SHARKEY, University of Sheffield (UK) Tim SMITHERS, CEIT (Spain) Tom ZIEMKE, University of Skovde (Sweden) Submissions to this special issue should be sent by March 15th, 1999 to: Carme Torras Institut de Robotica i Informatica Industrial (CSIC-UPC) Gran Capita 2-4 (edifici Nexus) 08034-Barcelona (Spain) e-mail: ctorras at iri.upc.es http://www-iri.upc.es/people/torras SCHEDULE: --------- March 15th, 99 - submission deadline May 15th, 99 - information to authors July 1st, 99 - deadline for final papers October 99 - publication of the special issue *** CONNECTION SCIENCE *** Journal of Neural Computing, Artificial Intelligence and Cognitive Research http://www.carfax.co.uk/cos-ad.htm =============================================================================== From A.van.Ooyen at nih.knaw.nl Fri Feb 26 11:24:01 1999 From: A.van.Ooyen at nih.knaw.nl (Arjen van Ooyen) Date: Fri, 26 Feb 1999 17:24:01 +0100 Subject: Poly-and Mononeuronal Innervation in a Model of Neuromuscular Junction Message-ID: <36D6CAA1.240E@nih.knaw.nl> Poly- and Mononeuronal Innervation in a Model for the Development of Neuromuscular Connections A. van Ooyen & D. J. Willshaw J. Theor. Biol. (1999) 196: 495-511. Download: http://www.cns.ed.ac.uk/people/arjen/papers/polyneuronal_net.ps.gz or request reprint: A.van.Ooyen at nih.knaw.nl http://www.cns.ed.ac.uk/people/arjen/competition.html ABSTRACT In the normal development of connections between motor neurons and muscle fibres, an initial stage of polyneuronal innervation is followed by withdrawal of connections until each muscle fibre is innervated by a single axon. However, polyneuronal innervation has been found to persist after prolonged nerve conduction block, in spite of the resumption of normal neuromuscular activity. Here we analyse in detail a model proposed for the withdrawal of nerve connections in developing muscle, based on competition between nerve terminals. The model combines competition for a presynaptic resource with competition for a postsynaptic resource. Using bifurcation and phase space analysis, we show that polyneuronal innervation, as well as mononeuronal innervation, can be stable. The model accounts for the development of mononeuronal innervation as well as for persistent polyneuronal innervation after prolonged nerve conduction block, which appears as a consequence of the general competitive interactions operating during normal development. -- Arjen van Ooyen, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. email: A.van.Ooyen at nih.knaw.nl website: http://www.cns.ed.ac.uk/people/arjen.html phone: +31.20.5665483 fax: +31.20.6961006 From dorota at at.net.au Sun Feb 28 08:11:19 1999 From: dorota at at.net.au (Dorota Kieronska) Date: Sun, 28 Feb 1999 21:11:19 +0800 Subject: CFP: ICONIP'99 Message-ID: <51540563F42FD2119B8600002139404504F3F3@tower0.at.net.au> ************************************** * * * ICONIP'99 * * jointly with * * ANZIIS'99 and ANNES'99 and ACNN'99 * * * ************************************** The 6th International Conference on Neural Information Processing-- jointly with the 7th Australian and New Zealand International Conference on Intelligent Information Processing Systems, and the 5th New Zealand International Conference on Artificial Neural Networks and Expert Systems, and the 11th Australian Conference on Neural Networks. ************************ * * * 16-20 November, 1999 * * Perth, Australia * * * ************************ In 1999, the annual conference of the Asian Pacific Neural Network Assembly, ICONIP'99, will be held jointly with three other major international conferences in the Asian Pacific Region, from 16 to 20 November 1999 in Perth, Western Australia. The joint conference will have three parallel streams: ************************************************************ * * Stream1: Biological Neural Information Processing * * Stream2: Neural Information Processing and Soft Computing * * Stream3: Applied Techniques * * ************************************************************ TOPICS OF INTEREST Stream 1: Biological Neural Information Processing Neurobiological systems Cognition Cognitive models of the brain Dynamical modelling, chaotic processes in the brain Brain computers, biological computers Consciousness, awareness, attention Adaptive biological systems Modelling emotions Perception, vision Learning languages Stream 2: Neural Information Processing and Soft Computing Soft computing--paradigms, methods, tools Artificial neural network models, architectures and algorithms Evolutionary programming and genetic algorithms Fuzzy systems Optimisation Expert systems Artificial life Adaptive systems Approximate reasoning Hybrid and ensemble systems Distributed AI systems, agent-based systems Machine learning, data mining and intelligent databases Probabilistic and statistical methods Stream 3: Applied techniques and Intelligent information systems Intelligent information retrieval systems Pattern recognition and image processing Speech recognition and language processing Human-computer interfaces, web computing Robotics and mechatronics Information engineering Time-series prediction Intelligent control Virtual reality Application of intelligent information technologies in: engineering, process industries, law, finance and business, medicine and sport, geographic information systems, geology, data mining, environmental protection, ... ************************************************************************ HONORARY CHAIR Shun-Ichi Amari, Tokyo University GENERAL CONFERENCE CHAIR, Tamas (Tom) Gedeon, University of New South Wales GENERAL CONFERENCE CO-CHAIR Yianni Attikiouzel, University of Western Australia Marwan Jabri, University of Sydney Nikola Kasabov, University of Otago PROGRAM COMMITEE CO-CHAIRS K Fukushima (jp), T Gedeon, S Halgamuge (au), N Kasabov (nz), D Nauck (de) INTERNATIONAL PROGRAM COMMITTEE (TENTATIVE) T Adams (fj), D Aha (us), D Alahakoon (au), I Aleksander (uk), S Amari (jp), J Andreae (nz), Y Attikiouzel (au), J Austin (uk), J Baldwin (jp), P Bartlett (au), A Bastian, J Bezdek (us), R Bogner (au), B Bouchon-Meunier (fr), B Bruegge (de), A Bulsara (us), T Caelli (au), T Chen (cn), V Cherkassky (us), S Cho, S Cho (kr), A Cichocki (jp), I Cloete (za), R Coggins (au), G Coghill (nz), G Deboeck (us), J Diederich, V Dimitrov (au), L Ding (sg), T Downs (au), W Duch (pl), W Dunin-Barkowski (su), R Eckmiller (de), F Esteva (es), E Falkenauer (be), M Fedrizzi (it), B Flower (au), D Fogel (us), W Friedrich (nz), K Fukushima, T Furuhashi (jp), T Gedeon (au), M Glesner (de), J Grantner (us), S Halgamuge (au), M Hassoun (us), H Hellendoorn (nl), T Hendtlass (au), K Hirota (jp), A Hoffmann, J Hughes (au), M Ishikawa (jp), M Jabri, I Jagielska, L Jain (au), M Jamshidi (us), A Jennings (au), S Jones (uk), J Kacprzyk (pl), V Karri (au), N Kasabov (nz), O Katai, H Kawakami (jp), O Kaynak (tr), L Koczy, S Kovacs (hu), A Kowalczyk (au), W Kurutach (th), R Kozma (nz), R Kruse (de), D Lakov (bg), S Lee (kr), B Lees, D Levy (au), M Lim (sg), P Maheshwari (au), R Marks (us), F Masulli (it), G Matsumoto (jp), D McMichael (au), J Miller (nz), S Mitra (us), M Mizumoto (jp), N Morgan (us), M Mukaidono, T Musha, T Nagano (jp), D Nauck (de), M Negnevitsky (au), M Negoita (us), C Newton, C Latimer (au), R O'Shea (nz), E Oja (fi), Y Okabe (jp), N Pal, S Pal (in), M Palaniswami (au), G Pasi (it), L Patnaik (in), M Paulin (nz), D Pham (uk), A Ralescu (tr), B Reusch (de), L Reznik (au), D Ruan (be), I Rudas (hu), P Sallis (nz), E Sanchez (fr), D Saffen, Y Sato (jp), B Schuermann (de), O Simula (fi), U Srinivasan (au), S Sugiyama (jp), R Sun, H Szu (us), M Takacs (yu), H Takagi (jp), J Taylor (uk), A Topchy (ru), P Treleavan (uk), N Tschiold-Guerman (ch), A Tsoi (au), Y Tsujimura, M Tsukada, E Uchino, S Usui (jp), J Herik (nl), V Vapnik, R Vemuri (us), N Vojdani, W Wahlster (de), A Waibel (us), P Wallis, J Wiles, W Wilson, P Wong (au), L Xu (hk), T Yamakawa, T Yanaru (jp), X Yao (au), D Yun, L Zadeh, J Zurada (us) ORGANISING COMMITTEE CO-CHAIRS: Patrick Wong (Chair), Tom Gedeon, Graham Mann, University of NSW CONFERENCE ORGANISER Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au CALL FOR PAPERS Papers must be received by 16 April 1999. They will be reviewed by senior researchers in the field and the authors will be informed about the decision of the review process in July 1999. The accepted papers must be submitted in a camera-ready format by 31 August. All accepted papers will be published by IEEE Computer Society Press. As the conference is a multi-disciplinary meeting the papers are required to be comprehensible to a wider rather than to a very specialised audience. Papers will be presented at the conference either in an oral or in a poster session. Please submit five copies of the paper written in English on A4-format white paper with 2.5 cm margins on all four sides, in two column format, on not more than 6 pages, single-spaced, in Times or similar font of 10 points, and printed on one side of the page only. Centred at the top of the first page should be the complete title, author(s), mailing and e-mailing addresses, followed by an abstract and the text. In the covering letter the stream and the topic of the paper according to the list above should be indicated. The IEEE Transaction journals LaTex article style can be used. SPECIAL ISSUES OF JOURNALS AND EDITED VOLUMES Selected papers will be published in special issues of scientific journals. The organising committee is looking for publications of edited volumes which include chapters covering the conference topics written by invited conference participants. TUTORIALS (16 November) Conference tutorials will be organized to introduce the basics of cognitive modelling, dynamical systems, neural networks, fuzzy systems, evolutionary programming, soft computing, expert systems, hybrid systems, and adaptive systems. Proposals for tutorials are due on 15 March 1999. EXHIBITION Companies and university research laboratories are encouraged to exhibit their developed or distributing software and hardware systems. STUDENT SESSION Postgraduate students are encouraged to submit papers to this session following the same formal requirements for paper submission. The submitted papers will be published in a separate brochure. VENUE (Perth) The Conference will be held in Perth city. The closing session will be held on Saturday, 20 November at a scenic bush location following an outdoor seafood barbecue. ACCOMMODATION Accommodation has been booked at the conference venue, and in the numerous Hotels nearby suitable for a range of budgets. Please see the web site for details after the paper deadline date. POSTCONFERENCE EVENTS The ICONIP'99 post conference workshop and expo on "Future Directions for Intelligent Information Processing Systems" will be held in Dunedin, New Zealand from 22-23 November 1999. Please see the workshop homepage http://divcom.otago.ac.nz/infosci/kel/iconip99-workshop.htm. An affiliated post-conference workshop is the third Australia-Japan joint Workshop on Intelligent and Evolutionary Systems will be held in Canberra, Australia from 22-23 November 1999. Please see the workshop homepage http://www.cs.adfa.edu.au/conference/aj99/ IMPORTANT DATES Papers due: 16 April 1999 Proposals for tutorials: 1 March 1999 Notification of acceptance: 15 July 1999 Final camera-ready papers due: 31 August 1999 Registration of at least one author of a paper: 31 August 1999 Early registration: 31 August 1999 CONFERENCE CONTACTS, PAPER SUBMISSIONS, CONFERENCE INFORMATION, REGISTRATION FORMS Conference Secretariat ICONIP'99 Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au web: or by post to: Conference Secretariat ICONIP'99 ------------------ Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983 Australia ------------------------------------------------------------------------ From zhaoping at gatsby.ucl.ac.uk Mon Feb 1 06:49:40 1999 From: zhaoping at gatsby.ucl.ac.uk (Dr Zhaoping Li) Date: Mon, 1 Feb 1999 11:49:40 GMT Subject: Ph.D. studentships in Computational Neuroscience at Gatsby Unit Message-ID: <199902011149.LAA15879@vision.gatsby.ucl.ac.uk> Applications are invited for research studentships in the field of computational neuroscience tenable from September 1999. In particular, applications are welcome from candidates with an interest to pursue research in the area of high and low level vision and/or motor control. Applicants should have a strong analytical background, a good honours degree in a relevant subject, and a keen interest in neuroscience. Competitive studentships are available. To apply, email a statement of your research interest and a text-only version of your CV to Janice Hankes at admissions at gatsby.ucl.ac.uk by March 1 1999. Further details about the Gatsby Unit can be found at http://www.gatsby.ucl.ac.uk. From smagt at dlr.de Tue Feb 2 06:51:38 1999 From: smagt at dlr.de (Patrick van der Smagt) Date: Tue, 2 Feb 1999 12:51:38 +0100 (MET) Subject: Announcement: IEE Workshop on cerebellar control (London, Feb. 15) Message-ID: <199902021151.MAA07667@ilz.robotic> WORKSHOP ANNOUNCEMENT ===================== Self-learning robots III Brainstyle robotics: The cerebellum beyond function approximation http://www.robotic.dlr.de/Smagt/workshop/iee99.html A Workshop organised by Professional Group A9 (Neural computing) Monday 15 February 1999, Savoy Place, London, UK Patrick van der Smagt (Institute of Robotics and System Dynamics, DLR), workshop co-organiser Noel Sharkey (University of Sheffield), workshop co-organiser Throughout the years there has been extensive research in the structure and working of the cerebellum. The cerebellum is a part of the brain of vertebrates that is well understood in various aspects: The cerebellum is used for the spatial motor coordination of fast movements. High precision of the movement is an important issue. The structure of the cerebellum is well-understood: the cells as well as their connections are known. Also the connections to the other parts of the brain and spine are understood. The learning process is well-understood since 1992. The 'traditional' neural approach for robot control typically consists of some form of function approximation. We gather a large amount of learning samples, and try to represent those as well as possible with some (universal or specialized) approximator. The major differences lie there where the choice of approximator is concerned. Does the cerebellum do something different? Many special issues and workshops focusing on brain models for robot control reflect the increased attention of the development of cerebellar models. But what can we expect of such models in the future? Is the cerebellum is 'just' a function approximator, and can we mimic its functionality using other models? In this workshop we will focus on 'working' models of the cerebellum; how far have we come? Can real applications be expected in the near future? Can cerebellar models do better than others? Are brainstyle robotics something to expect in the near future? Programme 09.00 Registration and Coffee 09.30 Welcome 09.40 Overview of cerebellar control Patrick van der Smagt (Institute of Robotics and System Dynamics, DLR) 10.20 The cerebellum and visually controlled movements Chris Miall (University Laboratory of Physiology, Oxford) 11.00 The cerebellum as a neuronal machine David Willshaw (University of Edinburgh) 11.40 Recurrent disinhibition implements a pointer-map module between cerebellar basket and Purkinje cells Klaus Hepp (Institute of Theoretical Physics, ETH Z"urich) 12.20 Lunch 13.20 Computational Delays and Habits Guido Bugmann (University of Plymouth) 14.00 Learning to Reach Via Corrective Movements Andrew Barto (University of Massachusetts) 15.00 Tea 15.15 Cerebellar vs. stiffness control Pietro Morasso (Universit'a di Genova) 15.55 Panel discussion 16.30 Close To register for the above event, please contact: Events Office, IEE, Savoy Place, London WC2R 0BL, tel: +44 (0)171 240 1871 ext: 2205/6, fax: +44 (0)171 497 3633 or email: events at iee.org.uk -- dr Patrick van der Smagt phone +49 8153 281152, fax -34 DLR/Institute of Robotics and System Dynamics smagt at dlr.de P.O.Box 1116, 82230 Wessling, Germany http://www.robotic.de/Smagt/ From A.van.Ooyen at nih.knaw.nl Wed Feb 3 09:06:23 1999 From: A.van.Ooyen at nih.knaw.nl (Arjen van Ooyen) Date: Wed, 03 Feb 1999 15:06:23 +0100 Subject: Conference Announcement Message-ID: <36B857DF.3E9F@nih.knaw.nl> Dear Computational Neuroscientist, This mailing is to bring to your attention the International Conference on Theory and Mathematics in Biology and Medicine 1999, that will be held from Tuesday June 29 through Saturday July 3, 1999, in Amsterdam, the Netherlands. At this joint meeting of the European Society for Mathematical and Theoretical Biology (ESMTB), the Society for Mathematical Biology (SMB), and the Netherlands Society for Theoretical Biology (NVTB), a full day is dedicated to Computational Neuroscience. This day will comprise two plenary invited lectures, given by Prof David Willshaw (University of Edinburgh, United Kingdom) and by Prof Idan Segev (Hebrew University, Israel), 12 contributing lectures of 20 minutes duration, as well as a poster session. Please find all information of the conference at http://www.bio.vu.nl/tmbm99/ We hope that the opportunity to emphasize Computational Neuroscience in the broad field of Mathematical and Theoretical Biology as well as the opportunity to become informed about the progress in other interesting topics at this meeting will encourage you to participate and submit an abstract. ******************************************************* The deadline of Abstract Submission is 1st March, 1999. ******************************************************* Kind regards, and we look forward to meeting you this summer in Amsterdam, Jaap van Pelt, Arjen van Ooyen Netherlands Institute for Brain Research, Amsterdam, The Netherlands Organisers Session Computational Neuroscience Brief overview of the Computational Neuroscience Session and the rest of the conference: ---------------------------------------------------------------- Understanding the brain in health and disease is perhaps the greatest challenge for science in the next century. Because the brain is an extremely complex and organised system from the molecular up to the behavioural level, this goal can not be achieved without modelling and computational approaches. The Computational Neuroscience Session aims at emphasizing this role of Mathematical Modeling and Computer Simulation in Neuroscience. Participants are invited to present progress in understanding the structure, function and development the brain. ----------------------------------------------------------------- The full-day major topics of this conference include 1. Non-Linear Population Dynamics 2. Epidemiology of Infectious Diseases 3. Sequencing, Folding and Molecular Structure 4. Evolution, Game Theory and Adaptive Dynami 5. Computational Neuroscience 6. Immunology and Within-Host Dynamics of Pathogens The half-day minor topics of this conference include 1. Individual-Based Population Dynamics 2. Gene Networks 3. Pattern Formation and Morphogenesis 4. Education in Mathematical Biology 5. Parasites and Diseases in Wildlife 6. Metabolic Control Networks 7. Whole Heart Modelling 8. Metapopulation Dynamics 9. Toxicology 10. Particle Based Modelling 11. Networks, Cellular Signalling and Biological Rhythms 12. Classification Methodology 13. Miscellaneous ========================================================== -- Arjen van Ooyen, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. email: A.van.Ooyen at nih.knaw.nl website: http://www.cns.ed.ac.uk/people/arjen.html phone: +31.20.5665483 fax: +31.20.6961006 From john at dcs.rhbnc.ac.uk Wed Feb 3 12:34:49 1999 From: john at dcs.rhbnc.ac.uk (John Shawe-Taylor) Date: Wed, 3 Feb 1999 17:34:49 +0000 (GMT) Subject: Informal pre-EuroCOLT'99 Workshop on Kernel Methods In-Reply-To: <01BE0EFC.3AF90BA0.jkh@dcs.rhbnc.ac.uk> Message-ID: We are hosting a one day informal workshop on Sunday 28th March at Nordkirchen Castle, Germany, on the Sunday before the EuroCOLT'99 conference; particular interest of the organisers is the analysis of Kernel Methods and this will be one of the themes of the workshop. To find out more information and registration procedures please go to the web site: http://svm.first.gmd.de/eurocolt99/workshop.html Organizing Committee John Shawe-Taylor, john at dcs.rhbnc.ac.uk Alex J. Smola, smola at first.gmd.de Bernhard. Schoelkopf, bs at first.gmd.de Robert C. Williamson, Bob.Williamson at anu.edu.au From Zoubin at gatsby.ucl.ac.uk Thu Feb 4 08:17:22 1999 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 4 Feb 1999 13:17:22 +0000 (GMT) Subject: Gatsby Unit: PhD and Postdoc positions Message-ID: <199902041317.NAA18273@cajal.gatsby.ucl.ac.uk> University College London Gatsby Computational Neuroscience Unit Current vacancies: PhD and Post-doctoral positions ------------------------------ Post-doctoral and PhD Research Positions High and low level vision Applications are invited for a post-doctoral research fellowship and a PhD research studentship in the field of high and low level vision tenable from September 1999. The successful candidate/s will be supervised by Dr Zhaoping Li. Applicants should have a strong analytical background, and a keen interest in neuroscience. Competitive salaries and studentships are available. ------------------------------ PhD Research Studentship Computational and experimental motor control Applications are invited for a PhD research studentship in the field of computational and experimental motor control tenable from September 1999. The successful candidate will be supervised by Dr Zoubin Ghahramani. Applicants should have a strong analytical background, and a keen interest in neuroscience. Competitive salaries and studentships are available. ------------------------------ APPLICATION DEADLINE: Monday 1 March 1999 HOW TO APPLY: http://www.gatsby.ucl.ac.uk/vacancies.html EMAIL ENQUIRIES: admissions at gatsby.ucl.ac.uk (Janice Hankes) From horwitz at helix.nih.gov Fri Feb 5 13:40:05 1999 From: horwitz at helix.nih.gov (Barry Horwitz) Date: Fri, 5 Feb 1999 13:40:05 -0500 Subject: Available Postdoctoral Position Message-ID: Postdoctoral Fellowship in Neural Modeling of Human Functional Neuroimaging Data National Institute on Aging and National Institute on Deafness and Other Communication Disorders An entry-level postdoctoral fellowship under the supervision of Dr. Barry Horwitz is available immediately for developing and applying computational neuroscience modeling methods to in vivo human functional neuroimaging data, obtained from positron emission tomography (PET) and functional magnetic resonance imaging (fMRI). The goal is to understand the relation between functional neuroimaging data, with its low spatial and temporal resolution, and the underlying electrophysiological behavior of multiple interconnected neuronal populations. Knowledge of neural modeling techniques and programming experience are required. PET and fMRI facilities are available to implement and test hypotheses derived from the modeling. See Cerebral Cortex 8: 310-320 (1998) for an example of this approach. PhD or MD degree required. The position is in the Language Section, Voice, Speech and Language Branch, NIDCD, NIH, Bethesda, MD USA. For further information, contact: Dr. Barry Horwitz, Bldg. 10, Rm. 6C-414, National Institutes of Health, Bethesda, MD 20892, USA. Tel. 301-594-7755; FAX: 301-402-0074; Email: horwitz at helix.nih.gov. ----------------------------------------------------------------- Barry Horwitz, Ph.D. Tel. 301-594-7755 Bldg. 10, Rm. 6C-414 FAX 301-402-0074 MSC 1588 horwitz at helix.nih.gov Laboratory of Neurosciences National Institute on Aging National Institutes of Health Bethesda, MD 20892 USA From lendaris at sysc.pdx.edu Fri Feb 5 20:12:30 1999 From: lendaris at sysc.pdx.edu (George G. Lendaris) Date: Fri, 05 Feb 1999 17:12:30 -0800 Subject: Director Position: Systems Science Message-ID: <36BB96FD.F83AF184@sysc.pdx.edu> Director, Interdisciplinary Systems Science Ph.D. Program. Looking for a dynamic, creative individual to lead a unique established interdisciplinary Ph.D. program to international prominence, building on resurgent interest in areas such as complexity, chaos, learning and adaptation. The existing program at Portland State University involves a core component and relationships with a dozen academic units with a current enrollment of 145 students. The Systems Science core faculty currently comprises three tenure track and one adjunct faculty. In addition to the general duties associated with directing and administrating the Program, the Director is expected to 1) provide strong interaction with faculty and administration of participating department/schools; 2) cultivate collaborative research and teaching among the faculty in Systems Science, Social Sciences, Natural Sciences, Mathematics, Engineering, and Business; 3) teach as an additional core faculty member; 4) guide the establishment of a new Masters in Systems Science (currently in the approval process); 5) initiate and guide activities to expand funding for research and program development; and 6) direct the student recruiting efforts. Applicants should have an earned doctorate in Systems Science or related discipline and experience in an academic setting including teaching, research publications and grants. Additional information about the Program is available at http://www.sysc.pdx.edu/. Applicants should submit 1) A letter of application; 2) a vita; and 3) three names of reference to Nancy A. Perrin, Chair Search Committee, Portland State University, P.O. Box 751-SYSC, Portland, OR 97207. Review of applications will begin on Feb. 26, 1999, and will continue until position is filled. PSU is an affirmative action/equal opportunity institution. ---------------------------------------------------------------- From bebis at cs.unr.edu Sat Feb 6 18:38:47 1999 From: bebis at cs.unr.edu (George Bebis) Date: Sat, 6 Feb 1999 15:38:47 -0800 (PST) Subject: IEEE International Conference on Information, Intelligence and Systems Message-ID: <199902062338.PAA01323@crete.cs.unr.edu> -------------- next part -------------- First Call for Papers IEEE International Conference on Information, Intelligence and Systems (Symposia) November 1-3, 1999 Washington, DC Sponsored by : IEEE Computer Society In Cooperation with : AAAI Society; SMC Society; NN Society; IAPR Society; ACM Society PAMI TC; Virtual Intelligence TC; TAI Conference; BU-CIS Center, TUCrete and UCrete Conference includes tours to NIST or NASA and NIH or TIGR http://iciis99.cs.unr.edu/ ------------------------------------------------------------------------ The IEEE International Conference on Information, Intelligence, and Systems (ICIIS) offers theoretical and practical media for the constructive interaction among scientists and practitioners from different research fields (computers, mathematics, physics, biology, medicine, chemistry, engineering, etc) having as goals the development of methodologies and tools for the solution of complex problems in neuroscience and biology, automation and robotics, image, speech and natural languages, and their integration. The conference is organized as four simultaneous single-track symposia. General Symposia Chair: N.G.Bourbakis, BU, CIS & UC Email:Bourbaki at Binghamton.edu Phone:(607) 777-2165 ------------------------------------------------------------------------ IEEE Symposium on INTELLIGENCE in NEURAL and BIOLOGICAL SYSTEMS (INBS) SOME TOPICS: Biological Models, Computational Biology, Evolutionary Dynamics, Evolutionary Computation Models, Molecular Biology, DNA Sequence Processing, Genome Processes, DNA Topologies, Genome Mapping and Sequencing, Learning, Perception, Models of Neural Nets, Neuroscientific Models, etc. Program Chairs: J.Gattiker, LANL, and J. Wang, NJIT Phone: (505) 665-0604 & (973) 596-3396 Email: gatt at lanl.gov & jason at village.njit.edu Submit summary to: JR Gattiker TA3, SM43, MSF645, DP13S, XCM Los Alamos National Laboratory Los Alamos, NM 87544 ------------------------------------------------------------------------ IEEE Symposium on INTELLIGENCE in AUTOMATION and ROBOTICS (IAR) SOME TOPICS: Robot Path Planning, Multiple Robot Motion Planning, Autonomous Navigation, Arm/Leg Control, Robot Vision, Visual Tracking, Human-Robot Symbiosis, Telerobotics, Micro and Nano Robotics, Robot Cooperation, Assembly Strategies, Sensors, , Walking/Running Robots, Task Planning, Process Planning, Scheduling, Intelligent Control, etc. Program Chairs: L. Tsoukalas, Purdue U. and A.Tascillo, Ford Phone: (765) 494-0198 & (313) 845-7427 Email: tsoukala at helios.ecn.purdue.edu & atascill at ford.com Submit summary to: L. Tsoukalas 1290 Nuclear Engineering Building Purdue University West Lafayette, IN 47907-1290 ------------------------------------------------------------------------ IEEE Symposium on IMAGE, SPEECH, NATURAL LANGUAGE SYSTEMS (ISNLS) SOME TOPICS: Image (Coding, Compression, Encryption, Segmentation, Enhancement, Restoration, Skeletonization, Analysis, Morphology, Databases, Video, Pattern Recognition, OCR, Image Understanding & Interpretations, Mathematical Methods); Speech (Coding, Compression, Processing, Analysis, Synthesis, Recognition, Understanding); Natural Language (NL) Processing, Computational Linguistics, Document Processing, NL Translation, NL Understanding, Multimedia, etc. Papers on the technical details of the commercial (e.g. , consumer electronics) , scientific, and medical applications of the above are encouraged. Program Chairs: G. Bebis, U. Nevada, and S. Amer, USGS Phone: (702) 784-6463 & (605) 594-6864 Email: bebis at cs.unr.edu & samer at edcmail.cr.usgs.gov Submit summary to: George Bebis Dept of Computer Science University of Nevada Reno, NV 89557 ------------------------------------------------------------------------ IEEE Symposium on INFORMATION and INTELLIGENT AGENTS (IIA) SOME TOPICS: Agent models and architectures, Agent-oriented programming, Communication issues and protocols, Cooperation and coordination, Cooperative information systems, Conflict resolution and negotiation, Distributed search, Intelligent agents, Multiagent planning and learning, Practical applications (Enterprise integration, Feature interaction, Information gathering, Manufacturing, Software agents, Software engineering), Testbeds and development environments, User interface issues. Program Chairs: C. Koutsougeras, Tulane U., and S. Mertoguno, FUJITSU. Phone: (504) 862-3369 & (408) 922-9520 Email: ck at eecs.tulane.edu & jmertohu at fmi.fujitsu.com Submit summary to: Chris Koutsougeras Electrical Engineering and Computer Science Dept. Tulane University New Orleans, LA 70118 ------------------------------------------------------------------------ Information for Authors: Please send four copies of a paper or an extended summary (3-6 pages) describing the methodology and/or results by May 15, 1999 to the appropriate Program Chair's address above. Notification of acceptance (please provide email address or fax number) by June 30, 1999. Camera ready articles by August 15, 1999. Publication of Papers and Awards: This is a fully refereed conference. The proceedings will be published by IEEE Computer Society Press. A collection of best papers will be published in an archival Journal and an Award will be given to each best paper per Symposium. For further details or more up-to-date information please visit the conference WWW page at http://iciis99.cs.unr.edu/ Special Sessions: Proposals are invited for special sessions on any topic relevant to the symposium. If you are interested in organizing an special session please follow the procedure outlined below: 1. Submit a proposal (topic area) to one the appropriate program chair. 2. Upon approval, recruit authors for the session (2 hours, 4/5 papers) and review papers for appropriate topic/contents. 3. Session chair submits abstract *package* of author/title/extended abstracts to the appopriate program chair by May 30. 4. Editorial comments, and official acceptance of the session will be made by the program chair(s) by June 30. From barba at cvs.rochester.edu Mon Feb 8 10:04:41 1999 From: barba at cvs.rochester.edu (Barbara Arnold) Date: Mon, 8 Feb 1999 10:04:41 -0500 Subject: Fellowship program Message-ID: Summer Undergraduate Fellowship Program. The University of Rochester's Center for Visual Science is pleased to announce the 1999 Summer Undergraduate Research Program. Undergraduates in their junior year of baccalaureate studies who have an interest in neurobiological, optical, psychophysical or computational aspects of vision are invited to join a member of our diverse faculty for a summer of supervised laboratory training and participation in vision seminars and journal clubs. Our fellowships will cover travel expenses and provide a stipend of $1000/month for up to 4 months of continuous training that will culminate in participation in the Undergraduate Summer Workshop, Aug 6-9. Rochester, located near Lake Ontario in western New York, is an internationally recognized center for visual and imaging science and provides a scenic and natural setting for a variety of outdoor activities. Applicants will be selected based on academic achievement and a demonstration of interest in pursuing studies in visual science. For application and/or more information, please contact: Barbara Arnold University of Rochester Center for Visual Science RC Box 270270 Rochester, NY 14627-0270 716-275-8659 barba at cvs.rochester.edu.(www.cvs.rochester.edu) Acceptance to Fellowship Program considered upon receipt of application. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Barbara N. Arnold Administrator email: barba at cvs.rochester.edu Center for Visual Science phone: 716 275 8659 University of Rochester fax: 716 271 3043 Meliora Hall 274 Rochester NY 14627-0270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From giles at research.nj.nec.com Mon Feb 8 10:01:01 1999 From: giles at research.nj.nec.com (Lee Giles) Date: Mon, 8 Feb 1999 10:01:01 -0500 (EST) Subject: New paper: neural network classification and prior class probabilities Message-ID: <199902081502.KAA00263@alta> A non-text attachment was scrubbed... Name: not available Type: text Size: 2990 bytes Desc: not available Url : https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/00000000/8b0ffe11/attachment-0001.ksh From avrama at gmu.edu Mon Feb 8 18:20:33 1999 From: avrama at gmu.edu (avrama) Date: Mon, 8 Feb 1999 18:20:33 -0500 (EST) Subject: PhD Fellowship Message-ID: George Mason University Institute for Computational Sciences and Informatics Ph.D. Research Position ------------------------------ Applications are invited to the Ph.D. program in Bioinformatics, Computational Neuroscience concentration. Fellowship money is available to work with Dr. Blackwell, a CSI faculty member and affiliated with the Krasnow Institute. The area of research is the biophysical basis of long term memory. Applicants are expected to have experience in C programming and UNIX operating systems; completed course work on Ordinary differential equations and linear algebra; and an interest in neuroscience. Applications may be downloaded from http://csi.gmu.edu/CSI/applications.html Information about CSI and the Krasnow Institute is available at http://csi.gmu.edu/ and http://www.gmu.edu/departments/krasnow/ Application deadline for the fall 1999 term is March 1 1999. From zhang at salk.edu Mon Feb 8 18:24:28 1999 From: zhang at salk.edu (Kechen Zhang) Date: Mon, 8 Feb 1999 15:24:28 -0800 (PST) Subject: Preprint available Message-ID: The following paper is available in PostScript form: Journal of Neuroscience, in press A theory of geometric constraints on neural activity for natural three-dimensional movement Kechen Zhang and Terrence J. Sejnowski Howard Hughes Medical Institute, The Salk Institute, and University of California, San Diego Abstract: Although the orientation of an arm in space or the static view of an object may be represented by a population of neurons in complex ways, how these variables change with movement often follows simple linear rules, reflecting the underlying geometric constraints in the physical world. A theoretical analysis is presented for how such constraints affect the average firing rates of sensory and motor neurons during natural movements with low degrees of freedom, such as a limb movement and rigid object motion. When applied to non-rigid reaching arm movements, the linear theory accounts for cosine directional tuning with linear speed modulation, predicts a curl-free spatial distribution of preferred directions, and also explains why the instantaneous motion of the hand can be recovered from the neural population activity. For three-dimensional motion of a rigid object, the theory predicts that, to a first approximation, the response of a sensory neuron should have a preferred translational direction and a preferred rotation axis in space, both with cosine tuning functions modulated multiplicatively by speed and angular speed, respectively. Some known tuning properties of motion-sensitive neurons follow as special cases. Acceleration tuning and nonlinear speed modulation are considered in an extension of the linear theory. This general approach provides a principled method to derive mechanism-insensitive neuronal properties by exploiting the inherently low dimensionality of natural movements. http://www.cnl.salk.edu/~zhang/objectwfigs.ps.Z 31 pages, compressed PostScript file about 2.0 Mb (7.7 Mb uncompressed) From Marc.VanHulle at med.kuleuven.ac.be Tue Feb 9 04:45:50 1999 From: Marc.VanHulle at med.kuleuven.ac.be (Marc Van Hulle) Date: Tue, 9 Feb 1999 10:45:50 +0100 Subject: NNSP'99 deadline extended to February 20 Message-ID: <199902090945.KAA21408@simone.neuro.kuleuven.ac.be> ************************************************ **** FINAL CALL FOR PAPERS **** **** **** **** Note extended submission deadline **** **** Submission deadline: February 20, 1999 **** ************************************************ 1999 IEEE Workshop on Neural Networks for Signal Processing August 23-25, 1999, Madison, Wisconsin NNSP'99 homepage: http://eivind.imm.dtu.dk/nnsp99 Thanks to the sponsorship of IEEE Signal Processing Society the ninth of a series of IEEE workshops on Neural Networks for Signal Processing will be held at the Concourse Hotel, Madison, Wisconsin, USA. The workshop will feature keynote addresses, technical presentations, panel discussions and special sessions. Papers are solicited for, but not limited to, the following areas: Paradigms: Artificial neural networks, support vector machines, Markov models, graphical models, dynamical systems, evolutionary computation, nonlinear signal processing, and wavelets. Application Areas: Image/speech/multimedia processing, intelligent human computer interfaces, intelligent agents, blind source separation, OCR, robotics, adaptive filtering, communications, sensors, system identification, issues related to RWC, and other general signal processing and pattern recognition. Theories: Generalization, design algorithms, optimization, parameter estimation, and network architectures. Implementations: Parallel and distributed implementation, hardware design, and other general implementation technologies. SPECIAL SESSIONS The workshop features special sessions on * Support vector machines * Intelligent human computer interfaces PAPER SUBMISSION PROCEDURE Prospective authors are invited to submit extended summaries of no more than 6 pages using the electronic submission procedure described at the workshop homepage http://eivind.imm.dtu.dk/nnsp99 Exceptionally, in case of mail submission, prepare 5 copies of extended summaries of no more than 6 pages. The top of the first page of the summary should include a title, authors' names, affiliations, address, telephone and fax numbers and email address. Please send paper submissions to: Jan Larsen NNSP'99, Department of Mathematical Modelling, Building 321 Technical University of Denmark DK-2800 Lyngby, Denmark Camera-ready full papers of accepted proposals will be published in a hard-bound volume by IEEE and distributed at the workshop. SCHEDULE Submission of extended summary: February 20, 1999 Notification of acceptance: March 31, 1999 Submission of photo-ready accepted paper: April 29, 1999 Advanced registration, before: June 30, 1999 ORGANIZATION General Chair Yu Hen HU University of Wisonsin-Madison email: hu at ece.wisc.edu Finance Chair Tulay ADALI University of Maryland Baltimore County email: adali at umbc.edu Proceedings Chair Elizabeth J. WILSON Raytheon Co. email: bwilson at ed.ray.com Proceedings Co-Chair Scott C. DOUGLAS Southern Methodist University email: douglas at seas.smu.edu Publicity Chair Marc van HULLE Katholieke Universiteit Leuven email: marc at neuro.kuleuven.ac.be From Michael_Tarr at brown.edu Tue Feb 9 11:36:28 1999 From: Michael_Tarr at brown.edu (Michael J. Tarr) Date: Tue, 9 Feb 1999 11:36:28 -0500 Subject: No subject Message-ID: For immediate posting: VISITING POSITION IN HUMAN PERCEPTION/COGNITION, BROWN UNIVERSITY. The Department of Cognitive and Linguistic Sciences at Brown University invites applications for a three-year visiting assistant professor position, to begin July 1, 1999 (non-tenure-track). Candidates should have an active research program in the area of perception, action, visual cognition, and/or attention, and would be expected to teach three courses per year. All applicants must have received the Ph.D. degree or equivalent by the beginning of the appointment. The initial deadline for applications is March 15, 1999, but applications will be accepted after that time until the position is filled. Please send CV, recent publications, a cover letter describing teaching and research interests, and three letters of reference to: Perception/Cognition Search Committee, Department of Cognitive and Linguistic Sciences, Box 1978, Brown University, Providence, RI 02912. Brown is an Equal Opportunity/Affirmative Action employer. Women and minorities are especially encouraged to apply. From sschaal at usc.edu Tue Feb 9 12:10:49 1999 From: sschaal at usc.edu (Stefan Schaal) Date: Tue, 9 Feb 1999 09:10:49 -0800 Subject: Postdoc position in Computational Motor Control Message-ID: <199902091710.JAA15291@rubens.usc.edu.> Postdoctoral Position in Computational Motor Control ----------------------------------------------------- The Computational Learning and Motor Control Laboratory at the University of Southern California, which is part of the Kawato Dynamic Brain Project at ATR in Japan, is looking for an outstanding postdoctoral fellow in Computational Motor Control. The position is available starting from April 99, initially for one year, but can be renewed until Sept. 2001. Research topics of our group are strongly interdisciplinary. On the one hand, we are interested in developing formal theories of how the mammalian brain solves problems of sensory motor control. On the other hand, we also try to test these theories with humanoid robots, ranging from human-arm-like robots up to a full-body humanoid with a complete oculomotor and vestibular system. We also employ behavioral experiments to compare human performance against robot performance and our developed theories. A strong emphasis of our research lies on developing statistical learning mechanisms for the self-organization of motor control, for instance, applied to the learning of internal models, the self-organization of internal representations, or high-level topics like imitation learning. More information can be found at: http://www-slab.usc.edu or http://www.erato.atr.co.jp The ideal candidate for this position has a strong background in statistical learning and neural networks, in computational neuroscience of motor control, in behavioral motor control, as well as in dealing with experimental and computer equipment. The candidate is expected to independently lead research projects, supervise students, and interact with our Japan-based part of the project through visits to Japan. The postdoctoral position allows a great deal of flexibility concerning the research topic as long as it fits within the broad range of our group's interests in computational motor control. Before sending any formal applications, including a CV and at least 3 letters of recommendation, send a (brief!!) email letter of intent to sschaal at usc.edu (Dr. Stefan Schaal), briefly describing your past and current research interests, and how you would imagine to integrate in our projects. Stefan Schaal Assistant Professor Computer Science and Neuroscience University of Southern California, HNB 103 Los Angeles, CA 90089-2520 & Head, Computational Learning Group Kawato Dynamic Brain Project (ERATO/JST) ATR, Japan phone: (213) 740 9418 fax: (213) 740 1510 From ckiw at dai.ed.ac.uk Tue Feb 9 13:07:17 1999 From: ckiw at dai.ed.ac.uk (Chris Williams) Date: Tue, 9 Feb 1999 18:07:17 +0000 (GMT) Subject: ICANN*99 Call for Workshop Proposals Message-ID: CALL FOR PROPOSALS ICANN*99 Post Conference Workshops 11 September 1999 Edinburgh, Scotland Workshops on current topics in the fields of research covered by ICANN99 will be held on Saturday 11 September 1999 immediately after the regular ICANN conference program. Proposals by qualified individuals interested in chairing one of these workshops are solicited. Topics should fall within the broadly defined area of neural networks research, including: Theory and Algorithms; Neurobiology and Computational Neuroscience; Cognitive Modelling; Industrial, Commercial and Medical Applications; Hardware and Neuromorphic Engineering; Control, Robotics and Adaptive Behaviour. The goal of the workshops is to provide an informal forum for researchers to discuss important issues of current interest. The workshops will be held on Saturday 11th September, the day following the conference, making it convenient for ICANN participants to stay in Edinburgh over a Saturday night and thus be eligible for cheaper flights. It is anticipated that the registration fee for the workshops will be nominal or perhaps zero. Please note that there is no funding available for the workshops. Focussed open and/or controversial issues are encouraged and preferred as workshop topics. Representation of alternative viewpoints and panel-style discussions are particularly encouraged. Workshop organizers will have responsibilities including: 1) coordinating workshop participation and content, which could involve, for example, arranging short informal presentations by experts working in an area, arranging for expert commentators to sit on a discussion panel and formulating a set of discussion topics, etc. 2) moderating or leading the discussion and reporting its high points, findings, and conclusions to the group during evening plenary sessions 3) writing a brief summary and/or coordinating submitted material for post-conference electronic dissemination. This call for proposals can be found at http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/icann/ and further information will be added as it becomes available. Submission Instructions ----------------------- Interested parties should submit via e-mail a short proposal for a workshop of interest by 29 March, 1999 to Dr Chris Williams (ckiw at dai.ed.ac.uk), Institute for Adaptive and Neural Computation, University of Edinburgh. Proposals should include a title, a description of what the workshop is to address and accomplish, the planned format (mini-conference, panel discussion, or group discussion, combinations of the above, etc), and the proposed number of speakers. Where possible, please also indicate potential invitees (particularly for panel discussions). We strongly encourage that the organizers reserve a significant portion of time for open discussion. 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. In addition, please send a brief resume of the prospective workshop chair, a list of publications, and evidence of scholarship in the field of interest. Submissions should include contact name, address, e-mail address, phone number and fax number if available. Proposals should be mailed electronically to ckiw at dai.ed.ac.uk. All proposals must be RECEIVED by March 29, 1999. If e-mail is unavailable, mail so as to arrive by the deadline to: ICANN*99 Workshops c/o Dr Chris Williams Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 5 Forrest Hill, Edinburgh EH1 2QL, Scotland, UK PROPOSALS MUST BE RECEIVED BY MARCH 29 1999 -Please Post- -------------------------------------------------------------------------- Chris Williams ckiw at dai.ed.ac.uk Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 5 Forrest Hill, Edinburgh EH1 2QL, Scotland, UK fax: +44 131 650 6899 tel: (direct) +44 131 651 1212 (department switchboard) +44 131 560 3090 http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/ From S.Holden at cs.ucl.ac.uk Thu Feb 11 08:07:50 1999 From: S.Holden at cs.ucl.ac.uk (Sean Holden) Date: Thu, 11 Feb 1999 13:07:50 +0000 Subject: Research Fellow post in Intelligent Data Analysis and Fusion Message-ID: <856.918738470@cs.ucl.ac.uk> University College London Department of Computer Science Research in Intelligent Data Analysis and Fusion A position is available for a Research Fellow to work in collaboration with SmithKline Beecham, Glaxo-Wellcome, Unilever and Zeneca on an EPSRC funded project selected by the NPL-Sira Faraday Partnership. The aim of the project is to develop a hybrid data mining, analysis and fusion system incorporating the most recent artificial intelligence, neural network, fuzzy logic, and genetic algorithm techniques. The research fellow will be responsible for development of a hybrid system tool, for day to day running of the project, for evaluating the performance of the system in collaboration with the above companies and NPL, and for overseeing technology transfer to Integral Systems Solutions Ltd, supplier of the Clementine data mining system. Applicants should have a strong background in artificial intelligence, computer science and the development of data mining systems. The Fellow will work in the Intelligent Systems Research Group with four research students on the project, one associated with each of the major collaborating companies. The Group has extensive experience in the development of hybrid intelligent systems, in particular for financial and commercial applications. Applicants must have good mathematical and computing skills with experience of data mining research or applications or have carried out research and development in a closely related area. The position is for up to three years. The Research Fellow will be on the University RA1A scale (17,869-21,331 pounds sterling, including London allowance, depending on experience and length of appointment). For more details of the post and project see: http://www.cs.ucl.ac.uk/staff/b.buxton/Faraday or e-mail at B.Buxton at cs.ucl.ac.uk. or tel: 0171-380-7294 Please send 3 copies of your CV (including the names of two referees) to: M Sutcliffe, Department of Computer science, University College London, Gower Street, WC1E 6BT. or e-mail: vacancy at cs.ucl.ac.uk. The closing date for applications is 3 March 1999. ************* From rcs at cogsci.ed.ac.uk Thu Feb 11 06:15:15 1999 From: rcs at cogsci.ed.ac.uk (rcs@cogsci.ed.ac.uk) Date: Thu, 11 Feb 1999 11:15:15 GMT Subject: MSc in Cognitive SCience and Natural Language Message-ID: <21892.199902111115@burns.cogsci.ed.ac.uk> ************************************************************************** ************************************************************************** MSc in "Cognitive Science and Natural Language" in the School of Cognitive Science, Division of Informatics, University of Edinburgh The new School of Cognitive Science, in the newly created Division of Informatics, takes over the teaching and research training activities of the former Centre for Cognitive Science. In an act of fundamental restructuring, the Centre for Cognitive Science and the Human Communication Research Centre, as well as the departments of Computer Science and Artificial Intelligence, have all merged to form the Division of Informatics. In very broad terms, Informatics is devoted to studying the structure, behaviour and design of computational, cognitive and communicating systems, both natural and artificial. TEACHING STAFF Mark Steedman Head of School computational linguistics, grammar formalisms, knowledge representation Chris Brew corpora, data intensive linguistics, language technology Jo Calder grammar formalisms, computational linguistics Bruce Graham computational neuroscience, neural networks Alexander Holt natural language semantics, computational linguistics Ewan Klein linguistic theory, phonology Alex Lascarides lexical and discourse processing, semantics, pragmatics Chris Mellish natural language processing and generation Paul Schweizer philosophical logic, philosophy of mind, philosophy of language Richard Shillcock psycholinguistics, cognitive modelling, cognitive neuropsychology Keith Stenning human memory, inference, connectionism Geraint Wiggins logic programming, intelligent music systems Roberto Zamparelli theoretical linguistics MSc POSTGRADUATE PROGRAMME The School of Cognitive Science directly continues the educational heritage of the Centre for Cognitive Science, which established its postgraduate programme in 1978 and is widely recognized internationally as one of the best programmes of it's kind. The School of Cognitive Science offers a course of postgraduate study centred on language and cognition. STRUCTURE OF THE PROGRAMME The specific modules available vary slightly from year to year, but the following list of seven contributing disciplines provides an accurate indication of the general pattern: Cognitive Psychology Computational Linguistics Data Intensive Linguistics Formal Logic Logic Programming and Knowledge Representation Neural Computation Theoretical Linguistics The teaching is carried out over two ten-week terms, and many of the contributing subjects have both first and second term modules. Students are expected to take four modules per term for assessment. Between May and September, MSc students work on a dissertation or project. In addition to the first year taught course, there are a number of advanced courses available within the Division of Informatics for students working beyond the MSc level. Also, there is a rich diversity of research workshops and working groups in which PhD students are active participants. STUDYING IN EDINBURGH Edinburgh contains the largest concentration of expertise in Artificial Intelligence and Natural Language Processing in Europe. Students have access to that expertise, to Edinburgh's large copyright libraries, and within Cognitive Science, to a substantial offprint library. The School possesses extensive computing facilities based on a network of Sun workstations and Apple Macintoshes; access to Edinburgh's concurrent supercomputer and other central computing services is easily arranged. Edinburgh, `Athens of the North', is a beautiful Georgian city. Home of the International Festival, it is endowed with green parklands and lies close to the wildscape of the Scottish Highlands. Edinburgh restaurants offer a wide range of ethnic cuisines, and there is a variety of theatres, galleries, museums and cinemas. Applicants typically have undergraduate degrees in one of the participating areas or an appropriate joint honours degree. UK and EU students following the MSc and PhD courses are eligible to apply for studentships. SCS will advise all students concerning funding possibilities. SCS attracts studentships from a variety of UK and non-UK funding bodies. Contact: Mrs Judith Gordon or Miss E Kerse MSc Admissions PhD Admissions Division of Informatics Division of Informatics University of Edinburgh University of Edinburgh 5 Forrest Hill James Clerk Maxwell Bldg Edinburgh EH1 2QL Kings Buildings UK Edinburgh EH9 3JZ UK email: msc-admissions at inf.ed.ac.uk phd-admissions at inf.ed.ac.uk FOR DETAILED UP TO DATE INFORMATION ON THE SCHOOL OF COGNITIVE SCIENCE h t t p : / / w w w . c o g s c i . e d . a c . u k / ************************************************************************** ************************************************************************** From ASJagath at ntu.edu.sg Thu Feb 11 20:50:52 1999 From: ASJagath at ntu.edu.sg (Jagath C Rajapakse (Dr)) Date: Fri, 12 Feb 1999 09:50:52 +0800 Subject: Postdoc position available in brain image analysis Message-ID: <6665AC0C667ED11186E308002BB487E10366D0F7@exchange2> Postdoctoral Position Available in Brain Image Analysis A postdoctoral research fellow position is available to work under the supervision of Dr. Jagath C. Rajapakse in the School of Applied Science, Nanyang Technological University, Singapore. The fellow will involve in a joint research project between the school and Singapore General Hospital investigating analysis of functional MR time-series and segmentation of neuroanatomical structures in structural MR head scans. Knowledge in image processing and experience in programming in C++ are required. Candidate should have a Ph.D. in a related field. For further information, contact: Dr. Jagath Rajapakse, School of Applied Science, Nanyang Technological University, N4 Nanyang Avenue, Singapore 639798. Email: asjagath at ntu.edu.sg, Phone: +65 790 5802, Fax: +65 792 6559 From stefan.wermter at sunderland.ac.uk Fri Feb 12 13:46:28 1999 From: stefan.wermter at sunderland.ac.uk (Stefan Wermter) Date: Fri, 12 Feb 1999 18:46:28 +0000 Subject: Job: Research Network Assistant Message-ID: <36C47704.712E6D2F@sunderland.ac.uk> Part-time Research Network Assistant Applications are invited for a two year research network assistant position. You will support a new international research network in the areas of neural networks and computational neuroscience in the framework of the EPSRC project EmerNet. The exciting new project examines the possible benefits of neuroscience and brain architectures for computational neural networks in the future. The candidate will play a key role in the support of this international research network including set up of web server for the network, restricted web programming, design and maintenance of web pages, collection of material in computational neuroscience, organisational support and administration for international workshops and technical writing. Candidates should have a degree in a Computing/Neurobiology/ Neuroscience discipline and an interest in artificial neural networks and computational neuroscience. Good communication and technical writing skills are required and some software engineering experience is an advantage. There will be a possibility to register for PhD part-time if so desired. 18.5 hours per week 12,733 - 14,222 (pro-rata) Informal enquiries welcomed by Professor Dr Stefan Wermter, e-mail Stefan.Wermter at sunderland.ac.uk Ref No: CETS48 Closing Date: 25/2/99 Application forms and further particulars are available by telephoning extensions 3578, 2429, 2055, or 2425 or E-Mail employee.recruitment at sunderland.ac.uk quoting appropriate reference number. ******************************************** Professor Stefan Wermter Research Chair in Intelligent Systems University of Sunderland Centre of Informatics School of Computing, Engineering and Technology St Peters Way Sunderland SR6 0DD United Kingdom phone: +44 191 515 3279 fax: +44 191 515 2781 email: stefan.wermter at sunderland.ac.uk http://www.his.sunderland.ac.uk/ http://osiris.sunderland.ac.uk/~cs0stw/ ******************************************** From Ronan.Reilly at ucd.ie Fri Feb 12 22:12:40 1999 From: Ronan.Reilly at ucd.ie (Ronan Reilly) Date: Fri, 12 Feb 1999 19:12:40 -0800 Subject: Call for submissions - Cognitive Science Conference - University College Dublin Message-ID: <009301be56fe$b947b980$6a042b89@numwaan.ucd.ie> CALL FOR SUBMISSIONS "Cognitive Science for the New Millennium" University College Dublin May 16-17, 1999 http://www.cs.ucd.ie/cogsci Guest Speakers James L. McClelland, Carnegie-Mellon University Paul Smolensky, Johns Hopkins University To mark the launch in the academic year 1999/2000 of the MA/MSc programme in Cognitive Science at University College Dublin, a conference is being organised to celebrate and evaluate the achievements of the discipline. Conference submissions are invited that deal with all aspects of cognitive science. Submissions are especially encouraged that focus on an evaluation of its achievements and/or on an exploration of promising future research directions. Submissions should comprise a one page extended abstract. The conference will have a single stream, involving just oral presentations. We strongly encourage submissions by e-mail as ASCII or postscript attachments sent to Ronan.Reilly at ucd.ie , with subject "cogsci submission". Hardcopy submissions should include three copies of the abstract and be sent to the address below. All communications regarding submissions will be by e-mail, so provision of an e-mail address is essential. All submissions must be received on or before the 31st of March. For more details see the conference web page at http://www.cs.ucd.ie/cogsci. _________________________________ Ronan Reilly, PhD Department of Computer Science University College Dublin Belfield Dublin 4 voice: +353-1-706 2475 fax: +353-1-269 7262 email: Ronan.Reilly at ucd.ie web: http://www.cs.ucd.ie/staff/rreilly From mmp at ai.mit.edu Sat Feb 13 18:12:27 1999 From: mmp at ai.mit.edu (Marina Meila) Date: Sat, 13 Feb 99 18:12:27 EST Subject: PhD Thesis: "Learning with MIXTURES OF TREES" - Marina Meila - MIT Message-ID: <9902132312.AA13085@sinaia.mit.edu> Dear connectionists, my thesis on learning tree graphical models and mixtures thereof is now available via my web page: http://www.ai.mit.edu/~mmp or directly from ftp://ftp.ai.mit.edu/pub/mmp/thesis/thesis.ps.Z The thesis presents: * Density estimation in multidimensional domains and graphical models * Tree distributions and their properties * Mixtures of trees: definition, properties, an efficient learning algorithm * Learning mixtures of trees in the Bayesian framework * Very fast tree learning algorithms for sparse discrete data * A new approach to hidden variable discovery, an empirical method for estimating the description length of a mixture model, a novel independence test based on large deviation theory * Experiments: density estimation and classification with mixtures of trees Abstract and Table of Contents follow: --------------------------------------------------------------------------- Abstract One of the challenges of density estimation as it is used in machine learning is that usually the data are multivariate and often the dimensionality is large. Operating with joint distributions over multidimensional domains raises specific problems that are not encountered in the univariate case. Graphical models are representations of joint densities that are specifically tailored to address these problems. They take advantage of the (conditional) independencies between subsets of variables in the domain which they represent by means of a graph. When the graph is sparse, graphical models provide an excellent support for human intuition and allow for efficient inference algorithms. However, learning the underlying dependence graph from data is generally NP-hard. The purpose of this thesis is to propose and to study a class of models that admits tractable inference and learning algorithms yet is rich enough for practical applications. This class is the class of mixtures of trees models. Mixtures of trees inherit the excellent computational properties of tree distributions (themselves a subset of graphical models) but combine several trees in order to augment their modeling power, thereby going beyond the standard graphical model framework. The thesis demonstrates the performance of the mixture of trees in density estimation and classification tasks. In the same time it deepens the understanding of the properties of the tree distribution as a multivariate density model. Among others, it shows that the tree classifier implements an implicit variable selection mechanism. Learning mixtures of trees from data is a central subject of this thesis. The learning algorithm that is introduced here is based on the the EM and the Minimum Weight Spanning Tree algorithms and is quadratic in the dimension of the domain. This algorithm can serve as a tool for discovering hidden variables in a special but important class of models where, conditioned on the hidden variable, the dependencies between the observed variables become sparse. Finally, it is shown that in the case of sparse discrete data, the original learning algorithm can be transformed in an algorithm that is jointly subquadratic and that in simulations achieves speedups factors of up to a thousand. ---------------------------------------------------------------------------- Learning with Mixtures of Trees Marina Meila-Predoviciu Table of Contents * Cover, etc (postscript, 14 pages) o Abstract o Acknowledgements o Contents o List of Figures o List of Tables o Index of Algorithms * Introduction (postscript, 12 pages) o Density estimation in multidimensional domains o Introduction by example o Graphical models of conditional independence + Examples of established belief network classes + Advantages of graphical models + Structure learning in belief networks + Inference and decomposable models o Why, what and where? Goal, contributions and road map of the thesis + Contributions + The mixture of trees model + An efficient learning algorithm + An accelerated learning algorithm for sparse data + A top-down approach to hidden variable discovery + A road map for the reader * Trees and their properties (postscript, 14 pages) o Tree distributions o Inference, sampling and marginalization in a tree distribution + Inference + Marginalization + Sampling o Learning trees in the Maximum Likelihood framework + Problem formulation + Fitting a tree to a distribution + Solving the ML learning problem o Representation capabilities o Appendix: The Junction Tree algorithm for trees * Mixtures of trees (postscript, 12 pages) o Representation power of mixtures of trees o Basic operations with mixtures of trees + Marginalization + Inference + Sampling o Learning mixtures of trees in the ML framework + The basic algorithm + Running time and storage requirements + Learning mixtures of trees with shared structure + Remarks on the learning algorithms o Summary and related work * Learning mixtures of trees in the Bayesian framework (postscript, 9 pages) o MAP estimation by the EM algorithm o Decomposable priors for tree distributions + Decomposable priors over tree structures + Priors for tree parameters: the Dirichlet prior + The Dirichlet prior in natural coordinates + Dirichlet priors for trees and mixtures * Accelerating the tree learning algorithm (postscript, 22 pages) o Introduction o Assumptions o Accelerated CL algorithms + First idea: Comparing mutual informations between binary variables + Second idea: computing cooccurrences in a bipartite graph data representation + Putting it all together: the aCL-I algorithm and its data structures + Time and storage requirements + The aCL-II algorithm + Time and memory requirements for aCL-II o Generalization to discrete variables of arbitrary arity + Computing cooccurrences + Presorting mutual informations + A ``chain rule'' expression for the entropy of a discrete variable + The mutual information of two non-cooccurring variables o Using the aCL algorithms with EM o Decomposable priors and the aCL algorithm o Experiments o Concluding remarks o Appendix: Bounding the number of lists * An Approach to Hidden variable discovery (postscript, 21 pages) o Structure learning paradigms o The problem of variable partitioning o The tree H model o Variable partitioning in the general case + Outline of the procedure + Defining structure as simple explanation o Experiments + Experimental procedure + Experiments with tree H models + General H models o Approximating the description length of a model + Encoding a multinomial distribution o Model validation by independence testing + An alternative independence test + A threshold for mixtures + Validating graphical models with hidden variables o Discussion * Experimental results (postscript, 17 pages) o Recovering the structure o Density estimation experiments + Digits and digit pairs images + The ALARM network and data set o Classification with mixtures of trees + Using a mixture of trees as a classifier + The AUSTRALIAN data set + The MUSHROOM data set + The SPLICE data set. Classification and structure discovery + The single tree classifier as an automatic feature selector * Conclusion (postscript, 2 pages) * References (postscript, 5 pages) or (html) ---------------------------------------------------------------------------- From horn at neuron.tau.ac.il Sun Feb 14 03:26:32 1999 From: horn at neuron.tau.ac.il (David Horn) Date: Sun, 14 Feb 1999 10:26:32 +0200 (IST) Subject: Position in Functional Brain Imaging at Tel Aviv University Message-ID: TEL AVIV UNIVERSITY SENIOR ACADEMIC POSITION IN FUNCTIONAL BRAIN IMAGIMG ADAMS SUPER CENTER FOR BRAIN STUDIES Applications are invited for a senior academic position in functional brain imaging. The successful applicant should have experience in fMRI and will be expected to lead a research program in this field in collaboration with different disciplines throughout Tel Aviv University. The appointment procedure shall be according to the nomination regulations of Tel Aviv University. Departmental affiliation will be decided according to the academic background of the appointee. Applications including a CV, list of publications and a short summary of research interests should be sent by April 15th, 1999 to: Prof. David Horn, Adams Super Center for Brain Studies, Kaplun Bldg. room #428, Tel Aviv University, Tel Aviv, 69978, Israel. (Fax: ++972-3-6407932; e-mail: brainctr at neuron.tau.ac.il). Candidates should arrange to have three letters of reference transmitted directly to the above address. From esann at dice.ucl.ac.be Mon Feb 15 06:36:20 1999 From: esann at dice.ucl.ac.be (ESANN) Date: Mon, 15 Feb 1999 12:36:20 +0100 Subject: ESANN'99 - European Symposium on Artificial Neural Networks Message-ID: <002201be58d7$68aa3270$6f1f6882@natacha.dice.ucl.ac.be> ---------------------------------------------------- | | | ESANN'99 | | | | 7th European Symposium | | on Artificial Neural Networks | | | | Bruges (Belgium) - April 21-22-23, 1999 | | | | Preliminary programme | ---------------------------------------------------- The preliminary programme of the ESANN 99 conference is now available on the Web: http://www.dice.ucl.ac.be/esann For those of you who maintain WWW pages including lists of related ANN sites: we would appreciate if you could add the above URL to your list; thank you very much! We try as much as possible to avoid multiple sendings of this call for papers; however please apologize if you receive this e-mail twice, despite our precautions. For 7 years the ESANN conference has become a major event in the field of neural computation. ESANN is a human-size conference focusing on fundamental aspects of artificial neural networks (theory, models, algorithms, links with statistics, data analysis, biological background,...). The programme of the conference can be found at the URL http://www.dice.ucl.ac.be/esann, together with practical information about the conference venue, registration,... Other information can be obtained by sending an e-mail to esann at dice.ucl.ac.be . ===================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - Microelectronics Laboratory 3, pl. du Levant - B-1348 Louvain-la-Neuve - Belgium tel: +32 10 47 25 51 - fax: + 32 10 47 25 98 mailto:esann at dice.ucl.ac.be * Conference secretariat D facto conference services 27 rue du Laekenveld - B-1080 Brussels - Belgium tel: + 32 2 420 37 57 - fax: + 32 2 420 02 55 mailto:esann at dice.ucl.ac.be ===================================================== From j-patton at nwu.edu Mon Feb 15 13:55:52 1999 From: j-patton at nwu.edu (Jim Patton) Date: Mon, 15 Feb 1999 10:55:52 -0800 Subject: Conference Announcement & Call for Papers Message-ID: <4.1.19990215104618.00bcc3c0@merle.acns.nwu.edu> 3rd IMACS/IEEE International Multiconference on Circuits, Systems, Communications and Computers Athens, Greece, JULY 4-8, 1999 Special Session on Control in Biological Systems Session Organizer: Dr. Kamran Iqbal Biological systems comprise all living organism, in particular the vertebrates. There has been a growing interest in the study of such systems since the early work of Sherington (The integrative action of the nervous system,1906) . Control in biological systems takes many forms, e.g., control of voluntary movements, control of gait and posture, control of muscle tension and joint stiffness, control of temperature, control of blood pressure, control of respiratory system, etc. Biological control is exerted through commands originating in the central nervous system, and is assisted by spinal reflexes and a low level motor servo. The neuro-physiological controller can be termed as autonomous, self-organizing, self-evolving, and hierarchical. Biological systems are distinguished by their complex and diverse nature. Presence of nonlinear dynamics, multiple degrees-of-freedom, inter-segmental couplings, controller and actuator redundancies, multiple levels of feedback, finite feedback delays, and the distributed nature of parameters makes modelling of biological systems a daunting task. The complexity inherent in biological control systems has long challenged the researchers and often precluded the use of many a powerful tools otherwise available to control theorists. New and exciting additions to systems theory have often evolved from study of biological systems, e.g., artificial neural networks. The proposed session aims to explore the control issues relevant to the biological systems. A non-exhaustive list of topics includes the following: physiological and neuro-muscular system modeling, musculo-skeletal models, biokinematics, biodynamics, biomechanics, motor control, motor learning, neurological control of muscle activity, skeltomotor reflexes, control of limb movement, control of posture and locomotion, control of skilled movements, control of manipulative abilities, control of blood pressure, respiratory/ventilatory control, biomedical applications of automatic control, functional electrical and functional neuromuscular stimulation, evolutionary approaches in biological systems, trajectory control of movements, identification and computer simulation of biological systems. Instructions to the authors: Please submit the following items by March 15, 1999: #1. A one page abstract via e-mail to: k-iqbal at nwu.edu (See Sample) #2. PDF(*.pdf) & MS Word (*.doc) file(s) for the full paper(s) via email (See Sample) #3. The CONSENT TO PUBLISH & TRANSFER OF COPYRIGHT FORM completed and signed (Please download and print the Form and mail to address below): Please Note: Only original papers, written in English, which have not previously been published elsewhere will be accepted. Overview papers which will be very well organized and give an extended survey of results in their topic are particularly welcome. For all other questions please contact k-iqbal at nwu.edu or write to: ***************************************************************** Dr. Kamran Iqbal 645 N. Michigan Ave Suite 1100 Chicago, IL 60611 Ph: 312-503-6801 ***************************************************************** Benefits of Contributing to the IMAC/IEEE MultiConference *Have your paper published in the Proceedings that will be distributed to all conference attendees (including you!) *Have your paper published in the Luxurious Books of World Scientific (independently on the Proceedings) *Obtain international recognition in front of the IEEE - IMACS audience *Exchange knowledge with qualified industry professionals *Enhance your CV with two independent Publications (one in the Proceedings, one in the Post-Conference Luxurious Books) *Gain recognition as a leader from your peers in the Circuits/Systems/Communications/Computers Engineering *Have the opportunity to visit Greece and combine the conference with your summer vacations in sun-drenched Greek islands. From barba at cvs.rochester.edu Fri Feb 12 13:03:24 1999 From: barba at cvs.rochester.edu (Barbara Arnold) Date: Fri, 12 Feb 1999 13:03:24 -0500 Subject: Summer Workshop on Perception, Action and Cognition (for undergrads) Message-ID: The University of Rochester, Center for Visual Science invites undergraduates to participate in our 1999 Summer Workshop "Perception, Action and Cognition" to be held August 6-8, 1999. The workshop will include lectures, and poster sessions, and will offer an interactive forum between participants and our faculty. The workshop will focus on technological innovations in visual perception and cognition, sensory-motor coordination, computational modeling of neuronal circuits, and cortical physiology and development. Hands-on demos will introduce participants to the diverse methodology used in our laboratories. We will provide fellowships for travel and living expenses for over 20 students who demonstrate an interest in vision and cognitive neuroscience. We encourage applications from students in their junior year. Contact: Barbara Arnold, Center for Visual Science, Uniiv of Rochester, Rochester, NY 14627, 716-275-2459, Barba at cvs.rochester.edu ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Barbara N. Arnold Administrator email: barba at cvs.rochester.edu Center for Visual Science phone: 716 275 8659 University of Rochester fax: 716 271 3043 Meliora Hall 274 Rochester NY 14627-0270 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ From tom at cse.unsw.edu.au Wed Feb 17 00:26:31 1999 From: tom at cse.unsw.edu.au (Tom Gedeon) Date: Wed, 17 Feb 1999 16:26:31 +1100 (EST) Subject: ICONIP'99 Call for Papers Message-ID: <990217052631.12097@cse.unsw.edu.au> Subject: ICONIP/ANZIIS/ANNES/ACNN'99 Call for papers ************************************ * * * ICONIP'99 * * jointly with * * ANZIIS'99 and ANNES'99 and ACNN'99 * * * ************************************ The 6th International Conference on Neural Information Processing-- jointly with the 7th Australian and New Zealand International Conference on Intelligent Information Processing Systems, and the 5th New Zealand International Conference on Artificial Neural Networks and Expert Systems, and the 11th Australian Conference on Neural Networks. ********************** * * * 16-20 November, 1999 * * Perth, Australia * * * ********************** In 1999, the annual conference of the Asian Pacific Neural Network Assembly, ICONIP'99, will be held jointly with three other major international conferences in the Asian Pacific Region, from 16 to 20 November 1999 in Perth (note the new venue!), Western Australia. The joint conference will have three parallel streams: ************************************************************ * * * Stream1: Biological Neural Information Processing * * Stream2: Neural Information Processing and Soft Computing * * Stream3: Applied Techniques & Intelligent Info. Systems * * * ************************************************************ TOPICS OF INTEREST Stream1: Biological Neural Information Processing Neurobiological systems Cognition Cognitive models of the brain Dynamical modelling, chaotic processes in the brain Brain computers, biological computers Consciousness, awareness, attention Adaptive biological systems Modelling emotions Perception, vision Learning languages Stream2: Neural Information Processing and Soft Computing Soft computing--paradigms, methods, tools Artificial neural network models, architectures and algorithms Evolutionary programming and genetic algorithms Fuzzy systems Optimisation Expert systems Artificial life Adaptive systems Approximate reasoning Hybrid and ensemble systems Distributed AI systems, agent-based systems Machine learning, data mining and intelligent databases Probabilistic and statistical methods Stream3: Applied techniques and Intelligent information systems Intelligent information retrieval systems Pattern recognition and image processing Speech recognition and language processing Human-computer interfaces, web computing Robotics and mechatronics Information engineering Time-series prediction Intelligent control Virtual reality Application of intelligent information technologies in: engineering, process industries, law, finance and business, medicine and sport, geographic information systems, geology, data mining, environmental protection, ... ***************************************************************************** HONORARY CHAIR Shun-Ichi Amari, Tokyo University GENERAL CONFERENCE CHAIR , Tamas (Tom) Gedeon, University of New South Wales GENERAL CONFERENCE CO-CHAIR Yianni Attikiouzel, University of Western Australia Marwan Jabri, University of Sydney Nikola Kasabov, University of Otago PROGRAM COMMITEE CO-CHAIRS K Fukushima (jp), T Gedeon, S Halgamuge (au), N Kasabov (nz), D Nauck (de) INTERNATIONAL PROGRAM COMMITTEE (TENTATIVE) T Adams (fj), D Aha (us), D Alahakoon (au), I Aleksander (uk), S Amari (jp), J Andreae (nz), Y Attikiouzel (au), J Austin (uk), J Baldwin (jp), P Bartlett (au), A Bastian, J Bezdek (us), R Bogner (au), B Bouchon-Meunier (fr), B Bruegge (de), A Bulsara (us), T Caelli (au), T Chen (cn), V Cherkassky (us), S Cho, S Cho (kr), A Cichocki (jp), I Cloete (za), R Coggins (au), G Coghill (nz), G Deboeck (us), J Diederich, V Dimitrov (au), L Ding (sg), T Downs (au), W Duch (pl), W Dunin-Barkowski (su), R Eckmiller (de), F Esteva (es), E Falkenauer (be), M Fedrizzi (it), B Flower (au), D Fogel (us), W Friedrich (nz), K Fukushima, T Furuhashi (jp), T Gedeon (au), M Glesner (de), J Grantner (us), S Halgamuge (au), M Hassoun (us), H Hellendoorn (nl), T Hendtlass (au), K Hirota (jp), A Hoffmann, J Hughes (au), M Ishikawa (jp), M Jabri, I Jagielska, L Jain (au), M Jamshidi (us), A Jennings (au), S Jones (uk), J Kacprzyk (pl), V Karri (au), N Kasabov (nz), O Katai, H Kawakami (jp), O Kaynak (tr), L Koczy, S Kovacs (hu), A Kowalczyk (au), W Kurutach (th), R Kozma (nz), R Kruse (de), D Lakov (bg), S Lee (kr), B Lees, D Levy (au), M Lim (sg), P Maheshwari (au), R Marks (us), F Masulli (it), G Matsumoto (jp), D McMichael (au), J Miller (nz), S Mitra (us), M Mizumoto (jp), N Morgan (us), M Mukaidono, T Musha, T Nagano (jp), D Nauck (de), M Negnevitsky (au), M Negoita (us), C Newton, C Latimer (au), R O'Shea (nz), E Oja (fi), Y Okabe (jp), N Pal, S Pal (in), M Palaniswami (au), G Pasi (it), L Patnaik (in), M Paulin (nz), D Pham (uk), A Ralescu (tr), B Reusch (de), L Reznik (au), D Ruan (be), I Rudas (hu), P Sallis (nz), E Sanchez (fr), D Saffen, Y Sato (jp), B Schuermann (de), O Simula (fi), U Srinivasan (au), S Sugiyama (jp), R Sun, H Szu (us), M Takacs (yu), H Takagi (jp), J Taylor (uk), A Topchy (ru), P Treleavan (uk), N Tschiold-Guerman (ch), A Tsoi (au), Y Tsujimura, M Tsukada, E Uchino, S Usui (jp), J Herik (nl), V Vapnik, R Vemuri (us), N Vojdani, W Wahlster (de), A Waibel (us), P Wallis, J Wiles, W Wilson, P Wong (au), L Xu (hk), T Yamakawa, T Yanaru (jp), X Yao (au), D Yun, L Zadeh, J Zurada (us), ORGANISING COMMITTEE CO-CHAIRS: Patrick Wong (Chair), Tom Gedeon, Graham Mann CONFERENCE ORGANISER Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au CALL FOR PAPERS AND FORMAT OF SUBMISSION Papers must be received by 16 April 1999. They will be reviewed by senior researchers in the field and the authors will be informed about the decision of the review process by 30 June 1999. The accepted papers must be submitted in a camera-ready format by 31 August. All accepted papers will be published by IEEE Computer Society Press. As the conference is a multi-disciplinary meeting the papers are required to be comprehensible to a wider rather than to a very specialised audience. Papers will be presented at the conference either in an oral or in a poster session. Please submit five copies of the paper written in English on A4-format white paper with 2.5 cm margins on all four sides, in two column format, on not more than 6 pages, single-spaced, in Times or similar font of 10 points, and printed on one side of the page only. Centred at the top of the first page should be the complete title, author(s), mailing and e-mailing addresses, followed by an abstract and the text. In the covering letter the stream and the topic of the paper according to the list above should be indicated. The IEEE Transaction journals LaTex article style can be used. SPECIAL ISSUES OF JOURNALS AND EDITED VOLUMES Selected papers will be published in special issues of scientific journals. The organising committee is looking for publications of edited volumes which include chapters covering the conference topics written by invited conference participants. TUTORIALS (16 November) Conference tutorials will be organized to introduce the basics of cognitive modelling, dynamical systems, neural networks, fuzzy systems, evolutionary programming, soft computing, expert systems, hybrid systems, and adaptive systems. Proposals for tutorials are due on 1 March 1999. EXHIBITION Companies and university research laboratories are encouraged to exhibit their developed or distributing software and hardware systems. STUDENT SESSION Postgraduate students are encouraged to submit papers to this session following the same formal requirements for paper submission. The submitted papers will be published in a separate brochure. VENUE (Perth) The Conference will be held at the Langley Hotel in Perth city, near the Opera House. The closing session will be held on Saturday, 20 November as a scenic bush location following an outdoor seafood barbecue. ACCOMMODATION Accommodation has been booked at the conference venue, and in the numerous Hotels nearby suitable for a range of budgets. Please see the web site for details after the paper deadline date. TRAVELLING To be added. POSTCONFERENCE EVENTS The ICONIP'99 post conference workshop and expo on "Future Directions for Intelligent Information Processing Systems" will be held in Dunedin, New Zealand from 22-23 November 1999. Please see the workshop homepage "http://divcom.otago.ac.nz/infosci/kel/iconip99-workshop.htm". An affiliated post-conference workshop is the third Australia-Japan joint Workshop on Intelligent and Evolutionary Systems will be held in Canberra, Australia from 22-23 November 1999. Please see the workshop homepage "http://www.cs.adfa.edu.au/conference/aj99/" IMPORTANT DATES Papers due: 16 April 1999 Proposals for tutorials: 1 March 1999 Notification of acceptance: 30 June 1999 Final camera-ready papers due: 31 August 1999 Registration of at least one author of a paper: 31 August 1999 Early registration: 31 August 1999 CONFERENCE CONTACTS, PAPER SUBMISSIONS, CONFERENCE INFORMATION, REGISTRATION FORMS Conference Secretariat ICONIP'99 Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au web: http://www.at.net.au/iconip99 ------------------------------------------------------------------------ ICONIP'99 jointly with ANZIIS'99 and ANNES'99 and ACNN'99 TENTATIVE REGISTRATION PLEASE PRINT Title:________________________________ Surname:______________________________ First Name:___________________________ Position:_____________________________ Organisation:_________________________ Department:___________________________ Address:______________________________ ______________________________________ City:_________________________________ Country:______________________________ Phone:________________________________ Fax:__________________________________ Email:________________________________ Yes/No. Will you attend the conference? Yes/No. Will you submit a paper? Yes/No. Will you attend the closing session on the cruise? Yes/No. Would you like any further information? Please send a copy of this completed form by E-mail to iconip99 at at.net.au or by post to: Conference Secretariat ICONIP'99 ------------------ Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983 Australia ------------------------------------------------------------------------ From m.niranjan at dcs.shef.ac.uk Wed Feb 17 05:24:46 1999 From: m.niranjan at dcs.shef.ac.uk (Mahesan Niranjan) Date: Wed, 17 Feb 1999 10:24:46 GMT Subject: JobJobJob Message-ID: <199902171024.KAA08190@bayes.dcs.shef.ac.uk> +++++Job++++++++Job+++++++++Job++++++++Job++++++++++Job++++++++Job++++++++ Department of Computer Science, The University of Sheffield, England. Post-Doctoral Research Assistant and PhD Studentship in Neural Networks and Machine Learning. Applications are invited for a post-doctoral research assistant and a PhD student to work on an EPSRC funded project: ``Critical Evaluation of Bayesian Methods in Neural Networks''. Both positions are for three years and available for immediate start. The age-linked salary for the RA is in the range 15,735 - 20,107 pounds per annum; standard EPSRC rules apply to the studentship. A longer description of the project can be found http://www.dcs.shef.ac.uk/~niranjan/jobs/rafp.html To apply, an application package should be obtained from: Personnel Department The University of Sheffield Firth Court, Sheffield S10 2TN http://www.shef.ac.uk/jobs; Phone: 44 144 222 1631 Closing date 01 March 1999 For informal inquiries and any specific questions on the research, contact Email: M.Niranjan at dcs.shef.ac.uk; Phone: 44 114 222 1805 +++++Job++++++++Job+++++++++Job++++++++Job++++++++++Job++++++++Job++++++++ From pj+ at andrew.cmu.edu Wed Feb 17 15:28:05 1999 From: pj+ at andrew.cmu.edu (Patsy J Mccarthy) Date: Wed, 17 Feb 1999 15:28:05 -0500 (EST) Subject: faculty position in Mathematical Sciences at CMU Message-ID: Please send the following job announcement to your mailing list. CARNEGIE MELLON UNIVERSITY Department of Mathematical Sciences The Department of Mathematical Sciences at Carnegie Mellon University seeks candidates with research interests in mathematical/computational biology. Qualified candidates at all levels are encouraged to apply. We are particularly interested in researchers with interests in neural modelling, brain function, and genome mathematics. A successful candidate for this position would be expected to interact with allied scientists in other centers and departments at Carnegie Mellon. Applicants should send a vita, list of publications, and a statement describing current and planned research. Candidates should also arrange to have at least three letters of recommendation sent to: Appointments Committee, Department of Mathematical Sciences, Carnegie Mellon University, Pittsburgh, PA 15213. Applications will be accepted until the position is filled. Carnegie Mellon University is an Affirmative Action/Equal Opportunity Employer. ============================================================================= Patsy McCarthy | /\ /\ Assistant to the Dept. Head | =( o o )= Dept. of Math Sciences, Carnegie Mellon Univ | / ~ \ WeH 6109, Ext. 82545 | ( | | ) ~^ ^~} From uwe.zimmer at gmd.gr.jp Wed Feb 17 23:53:27 1999 From: uwe.zimmer at gmd.gr.jp (Uwe R. Zimmer) Date: Thu, 18 Feb 1999 13:53:27 +0900 Subject: PostDoc Pos at GMD-Japan RL (Distributed Systems) Message-ID: <36CB9C72.9581E5DC@gmd.gr.jp> PostDoc-Pos-Announcement -------------------------------------------------------- Post-Doctoral Research Positions in Distributed Systems -------------------------------------------------------- Two new post-doctoral positions are open at GMD Japan Research Laboratory, Kitakyushu, Japan and will be filled at the earliest convenience. The new laboratory (started in November '98 with a team of 8 scientists and 4 support staff, where these first 8 positions should be filled until April '99) is based on long term cooperations with the Japanese research community and focuses on the robotics and the telecooperation research fields. Investigated topics (in the area of telecooperation) are: - Distributed systems - Awareness in remote or of global situations - Remote operation - Non-verbal communications - Coupled virtual environments Computer science aspects of distributed systems, several application layers (currently open for discussion) as well as connected psychological constraints and effects need to be considered simultaniously (but not necessarely by every individual researcher). Although we are an independent research group, there are of course close connections to GMD in Germany (German National Research Center for Information Technology - http://www.gmd.de/). GMD Japan Research Laboratory is participating in the currently established Gigabit-network project in Japan. If the above challenges rose your interest, please proceed to our expectations regarding the ideal candidate: - Ph.D. / doctoral degree in computer sciences, electrical engineering, physics, mathematics, psychology or related disciplines. - Experiences in experimenting in distributed environments - Theoretical foundations in mathematics, control, protocols, coding, or systems theory. - Interest in joining an international team of motivated researchers. Furthermore it is expected that the candidate evolves/introduces her/his own perspective on the topic, and pushes the goals of the whole group at the same time. Salary starts at 8 Mill. Yen per year depending on experience. Support concerning the transfer to the Japanese environment is provided. For any further information, and applications (including addresses of referees, two recent publications, and a letter of interest!) please contact: Uwe R. Zimmer (address below) P.S. preferred formats for applications: pdf, ps.z, or FrameMaker ___________________________________________ ____________________________| Dr. Uwe R. Zimmer ___| GMD - Japan Research Laboratory | AIM Building 8F, 3-8-1, | Asano, Kokurakita-ku | Kitakyushu-city, 802-0001, Japan | _________________________________________________________________| Voice: +81 93 512 1566 - Fax: +81 93 512 1588 | http://www.gmd.gr.jp/ | From baveja at research.att.com Thu Feb 18 09:18:53 1999 From: baveja at research.att.com (satinder singh baveja) Date: Thu, 18 Feb 1999 09:18:53 -0500 Subject: Extended Deadline for MLJ Special issue on Reinforcement Learning... Message-ID: <36CC214C.CD483A76@research.att.com> Hello All, Given that the UAI conference deadline is Feb 21 and Colt is March 1st, I am extending the submission deadline for the Machine Learning Journal Special Issue on Reinforcement Learning by 15 days to March 15. Several folks have written to me requesting this. If you are going to submit - an early e-mail submission of title and abstract (even if it changes a bit in the final submission) will be much appreciated. best wishes, satinder singh -------------- next part -------------- With apologies for multiple postings.... --------------------------------------------------------------------------- Machine Learning Journal Special Issue on REINFORCEMENT LEARNING Edited by Satinder Singh **Extended ** Submission Deadline: March 15, 1999 Reinforcement learning has become an exciting focus for research in machine learning and artificial intelligence. Some of this new excitement comes from a convergence of computational approaches to planning and learning in large-scale environments, some from accumulating empirical evidence that reinforcement learning algorithms can solve significant real-world problems, and some from the ever-increasing mathematical maturity of this field. In this special issue we would like to capture a snapshot of this recent excitement, and invite submission of papers describing new work in reinforcement learning. New algorithms, problem frameworks, theoretical results, and empirical results are all appropriate contributions. Submission Deadline: March 15, 1999 **** Extended deadline **** Expected publication Date: January, 2000 SUBMISSION INSTRUCTIONS *********************** Submissions will undergo the standard Machine Learning journal review process. Send one hard-copy of submissions to: Satinder Singh AT&T Shannon Lab Room A269 180 Park Avenue Florham Park, NJ 07932 USA Fax: (973) 360-8970 E-mail: baveja at research.att.com If possible, e-mail the abstract separately to baveja at research.att.com, even before it is final, so that reviewers can be allocated efficiently. Also mail five hard-copies of submitted papers to: Karen Cullen MACHINE LEARNING Editorial Office Kluwer Academic Publishers 101 Philip Drive Assinippi Park Norwell, MA 02061 USA phone: (617) 871-6300 E-mail: karen at world.std.com **** PLEASE INDICATE that your submission is for the REINFORCEMENT **** LEARNING SPECIAL issue. Submissions exceptionally longer than 24 pages when formatted according the journal's style (pointers to which are below) may be rejected without review. Note: Machine Learning is now accepting submission for FINAL copy in electronic form. There is a latex style file and a related files available via the url: http://www.wkap.nl/journalhome.htm/0885-6125 Authors are strongly encouraged to use these style files or the formatting instructions stated at the back of the Machine Learning journal for their submissions. From mel at lnc.usc.edu Thu Feb 18 20:37:31 1999 From: mel at lnc.usc.edu (Bartlett Mel) Date: Thu, 18 Feb 1999 17:37:31 -0800 Subject: Preprint: Why Have Dendrites? Message-ID: <36CCC05B.B503E48C@lnc.usc.edu> The following preprint is now available: ========================================================= "Why Have Dendrites? A Computational Perspective" Bartlett W. Mel Biomedical Engineering Department University of Southern California ----- To appear in "Dendrites". G. Stuart, N. Spruston, & M. Hausser, (Eds.), Oxford University Press, 1999. Download: http://lnc.usc.edu/abstracts/99.mel_dendrites.html 18 pages, 431K gzipped postscript ----- ABSTRACT In most neurons, dendrites receive the vast bulk of the cell's synaptic input. But why do dendrites exist? What benefits do they confer upon a neuron? What computing operations are carried out in dendritic trees which could not occur within an electrically compact neuron? In this chapter we consider evidence from both physiological and computer modeling studies suggesting that dendrites provide space for a large number of compartmentalized nonlinear synaptic interactions. We thus narrow our sights to consider only those forms of dendritic integration in which spatial de-localization of synaptic inputs is a necessary feature of the postsynaptic computation. We illustrate the power and generality of this view of dendritic function by enumerating several diverse ways in which intradendritic computations could contribute to known nonlinear receptive field properties in visual cortex. -- Bartlett W. Mel (213)740-0334, -3397(lab) Assistant Professor of Biomedical Engineering (213)740-0343 fax University of Southern California, OHE 500 mel at lnc.usc.edu, http://lnc.usc.edu US Mail: BME Department, MC 1451, USC, Los Angeles, CA 90089 Fedex: 3650 McClintock Ave, 500 Olin Hall, LA, CA 90089 From sima at uivt.cas.cz Fri Feb 19 04:07:53 1999 From: sima at uivt.cas.cz (Jiri Sima) Date: Fri, 19 Feb 1999 10:07:53 +0100 Subject: New report available Message-ID: <36CD29E9.2440408A@uivt.cas.cz> Dear Colleagues, the following report Sima, J., Orponen, P. A Continuous-Time Hopfield Net Simulation of Discrete Neural Networks. Technical report V-773, ICS CAS, Prague, January, 1999. is now available on-line at http://www.uivt.cas.cz/ics/reports.html Abstract: We investigate the computational power of continuous-time symmetric Hopfield nets. Since the dynamics of such networks are governed by Liapunov (energy) functions, they cannot generate infinite nondamping oscillations, and hence cannot simulate arbitrary (potentially divergent) discrete computations. Nevertheless, we prove that any convergent fully parallel computation by a network of $n$ discrete-time binary neurons, with in general asymmetric interconnections, can be simulated by a symmetric continuous-time Hopfield net containing $14n+6$ units using the saturated-linear sigmoid activation function. In terms of standard discrete computation models this result implies that any polynomially space-bounded Turing machine can be simulated by a polynomially size-increasing sequence of continuous-time Hopfield nets. Similar techniques as here yield corresponding results on the convergence time and computational power of discrete-time Hopfield nets. Comments are welcome, Best regards, -- Jirka Sima ****************************************************************** Mgr. Jiri Sima, CSc. Department of Theoretical Informatics Institute of Computer Science, Czech Academy of Sciences Pod vodarenskou vezi 2, 182 07 Prague 8, Czech Republic phone: (+4202) 66 05 30 30 e-mail: sima at uivt.cas.cz fax: (+4202) 85 85 789 www: http://www.uivt.cas.cz/~sima ****************************************************************** From saadd at helios.aston.ac.uk Fri Feb 19 05:06:00 1999 From: saadd at helios.aston.ac.uk (saadd@helios.aston.ac.uk) Date: Fri, 19 Feb 1999 10:06:00 +0000 (GMT) Subject: ON-LINE LEARNING IN NEURAL NETWORKS Message-ID: <3166.199902191006@sun.aston.ac.uk> The following book is available from Cambridge University Press; see http://www.cup.cam.ac.uk/Scripts/webbook.asp?isbn=0521652634 ON-LINE LEARNING IN NEURAL NETWORKS David Saad, Ed. Series: Publications of the Newton Institute ISBN: 0 521 65263 4 DESCRIPTION On-line learning is one of the most powerful and commonly used techniques for training large layered networks, and has been used successfully in many real-world applications. Traditional analytical methods have been recently complemented by ones from statistical physics and Bayesian statistics. This powerful combination of analytical methods provides more insight and deeper understanding of existing algorithms, and leads to novel and principled proposals for their improvement. This book presents a coherent picture of the state-of-the-art in the theoretical analysis of on-line learning. An introduction relates the subject to other developments in neural networks and explains the overall picture. Surveys by leading experts in the field combine new and established material and enable non-experts to learn more about the techniques and methods used. This book, the first in the area, provides a comprehensive view of the subject and will be welcomed by mathematicians, scientists and engineers, whether in industry or academia. CONTENTS Foreward - Christopher M. Bishop 1. Introduction David Saad 2. On-line Learning and Stochastic Approximations Leon Bottou 3. Exact and Perturbative Solutions for the Ensemble Dynamics Todd K. Leen 4. A Statistical Study on On-line Learning Noboru Murata 5. On-line Learning in Switching and Drifting Environments with Application to Blind Source Separation Klaus-Robert Mueller, Andreas Ziehe, Noboru Murata and Shun-ichi Amari 6. Parameter Adaptation in Stochastic Optimization Luis B. Almeida, Thibault Langlois, Jose D. Amaral and Alexander Plakhov 7. Optimal On-line Learning in Multilayer Neural Networks David Saad and Magnus Rattray 8. Universal Asymptotics in Committe Machines with Tree Architecture Mauro Copelli and Nestor Caticha 9. Incorporating Curvature Information into On-line Learning Magnus Rattray and David Saad 10. Annealed On-line Learning in Multilayer Neural Networks Siegfried Boes and Shun-ichi Amari 11. On-line Learning of Prototypes and Principal Components Michael Biehl, Ansgar Freking, Matthias Hoelzer, Georg Reents and Enno Schloesser 12. On-line Learning whith Time-Correlated Examples Tom Heskes and Wim Wiegerinck 13. On-line Learning from Finite Training Sets David Barber and Peter Sollich 14. Dynamics of Supervised Learning with Restricted Training Sets Anthony C.C. Coolen and David Saad 15. On-line Learning of a Decision Boundary with and without Queries Yoshiyuki Kabashima and Shigeru Shinomoto 16. A Bayesian Approach to On-line Learning Manfred Opper 17. Optimal perceptron learning: an on-line Bayesian approach Sara A. Solla and Ole Winther See: http://www.ncrg.aston.ac.uk/books/OLNN/index.html From simon.schultz at anu.edu.au Fri Feb 19 17:40:35 1999 From: simon.schultz at anu.edu.au (Simon R. Schultz) Date: Sat, 20 Feb 1999 09:40:35 +1100 Subject: Preprint available Message-ID: <36CDE863.8F5D029B@anu.edu.au> The following paper has been accepted for publication in Proceedings of the Royal Society Series B: Biological Sciences, and is available in preprint form: "Correlations and the encoding of information in the nervous system", S. Panzeri*, S. R. Schultz*, A. Treves+ and E. T. Rolls* * Department of Experimental Psychology, University of Oxford, South Parks Rd., Oxford OX1 3UD, U.K. + SISSA - Programme in Neuroscience, 34013 Trieste, Italy ------ Abstract: ------ Is the information transmitted by an ensemble of neurons determined solely by the number of spikes fired by each cell, or do correlations in the emission of action potentials also play a significant role? We derive a simple formula which enables this question to be answered rigorously for short timescales. The formula quantifies the corrections to the instantaneous information rate which result from correlations in spike emission between pairs of neurons. The mutual information the ensemble of neurons conveys about external stimuli can thus be broken down into firing rate and correlation components. This analysis provides fundamental constraints upon the nature of information coding -- showing that over short timescales, correlations cannot dominate information representation, that stimulus-independent correlations may lead to synergy (where the neurons together convey more information than they would considered independently), but that only certain combinations of the different sources of correlation result in significant synergy rather than in redundancy or in negligible effects. This analysis leads to a new quantification procedure which is directly applicable to simultaneous multiple neuron recordings. This paper may be downloaded from the web page http://www.cogneuro.ox.ac.uk/~schultz/corrinfo.html, in either compressed postscript or PDF format. -- Simon Schultz, D.Phil. Visiting Fellow, Psychobiology Laboratory, Division of Psychology, Building 38, Australian National University, Canberra, ACT 0200, Australia. Phone: +61-2 6279 9733 Fax: +61-2 6249 0499 http://www.cogneuro.ox.ac.uk/~schultz From S.Singh at exeter.ac.uk Fri Feb 19 13:10:22 1999 From: S.Singh at exeter.ac.uk (Sameer Singh) Date: Fri, 19 Feb 1999 18:10:22 +0000 (GMT Standard Time) Subject: Lectureship at Exeter University, UK In-Reply-To: Message-ID: School of Engineering and Computer Science LECTURER IN COMPUTER SCIENCE VACANCY (ref number 6084) Further Particulars We are seeking an active, energetic researcher to contribute to one of the research areas in Computer Science: artificial intelligence, distributed and parallel systems, media computing, pattern recognition and neural computing. We would expect the successful candidate to contribute to the teaching programme at all levels and to develop modules in their own area of interest. This is a permanent position, available from the Summer Term 1999, or as soon as possible thereafter. The School The School of Engineering and Computer Science was formed on 1 August 1998 through the merger of the former Department of Computer Science and School of Engineering. For the purposes of student recruitment and quality assessment, the subject departments of Computer Science and Engineering retain an identity within the new School. For management and other purposes, the School is a fully integrated academic unit under the Head of School, Professor David Owens. School-level committees handle both research and teaching policy matters. The philosophy of the new School is to promote interdisciplinary work across the boundaries of engineering and computing and between the engineering sub-disciplines. The approach has been in place within engineering since the formation (by merger of separate departments) of the former School of Engineering in 1988. The merger with the Department of Computer Science, founded in 1979, represents an exciting opportunity to extend this concept at all levels: undergraduate, postgraduate and staff. One of our objectives is to establish a degree programme that bridges the two disciplines: a major/minor degree in Electronics with Computer Science. At present third year students in Computer Science regularly take options within Engineering. The School also runs full MSc, MPhil and PhD programmes. At present, Computer Science and Engineering are located in separate buildings on the main University campus, but plans are being developed to bring the whole School into single premises with the Harrison building (the current home of Engineering). The relocation will provide the School, by the summer of 2000, with new, extensive, purpose-built laboratory and computing facilities. Computer Science There are currently eleven academic staff in Computer Science. All academic staff are expected to contribute to research and teaching and to undertake appropriate administrative duties. A Computer Support Group of seven, including a hardware technician, maintains and upgrades the Department's computing facilities in conjunction with the University's IT Services. Members of the group are able also to provide support to specific research projects. In our major undergraduate laboratories we run Windows 98/95 and Linux. In support of these clients, clusters of Silicon Graphics workstations act as servers connected by Samba, a public domain Internet implementation of a Windows NT service. The research laboratories generally use Unix, either PCs running Linux, Silicon Graphics running Irix or Sun workstations running SunOS or Solaris. In total, we run 16 Silicon Graphics workstations connected by an ATM network capable of high speed video transmission, 11 Sun workstations, 50 Pentium PCs including 27 multimedia PCs running Windows 98/95 or Linux, 4 laser printers including a high quality colour laser printer. Supported applications include software for mathematical computing, 3D animation and visualisation, parallel and distributed systems, image processing and neural networks. All our computers are connected to the Internet. We run our own web server including an internal Intranet. Research groups have their own specialist servers connected to the network. The University's IT Services provides undergraduate accounts and filespace and other specialist facilities. We maintain a World Wide Web entry with home page: http://www.dcs.exeter.ac.uk Details of the syllabus for the current taught modules may be found under the syllabus entry on the page: http://www.dcs.exeter.ac.uk/wyska Research Computer Science was graded 4A in the 1996 Research Assessment Exercise. While recognising the importance of fostering individual initiative in research, the school encourages the formation of groups which act as a stimulus to research, enable informed but informal critical discussion and provide a supportive environment for postgraduate study. We currently recognise three research groups in Computer Science: Pattern Recognition and Neural Computing (Director: Dr Sameer Singh) Pattern Analysis and Neural Networks represent technologies of the future at the cutting edge of computer science research. This group investigates research issues in a number of applied topics in computer science and engineering. Our interests extend to Pattern Recognition, Neural Networks, Image Processing, Medical Imaging, OCR, Financial Forecasting and Linguistic Computing. A large number of our projects are with external collaborators and are supported by both internal and external grants. The group is committed to excellence in research achieved through a range of interactions at national and international level. At present the research group has more than twelve full-time members. Full details about the group are available at the web site: http://www.dcs.ex.ac.uk/research/pann Distributed and Parallel Systems (Director: Dr Stephen Turner) Research within this group concentrates on programming systems formed by connecting a number of processors, where the connection could be as tight as the bus of a multiprocessor, or as loose as the Internet. Our work on parallel computing involves harnessing many processors to solve a single problem, with discrete event simulation being an area of particular interest. As well as running parallel programs, we are interested in predicting their performance on different machine configurations. Another area of interest is the successful exploitation of distributed database technology, where we research into its application in solving extant problems in complex information sharing domains. Finally, we are also interested in programming networked embedded computing devices using the idea of mobile code. Artificial Intelligence (Director: Dr Antony Galton) The work of this research group comprises three main strands:- Knowledge Representation, with particular application to qualitative spatial and temporal reasoning; Media Computing, including natural language engineering and information retrieval, GIS and visualisation, multimedia databases, and psychological aspects of multimedia; Future Computing Paradigms, such as evolutionary computing and genetic algorithms, and quantum computing In accordance with our research strategy, special purpose laboratories have been established with dedicated equipment within the first two of these sub-areas. We also maintain a Media Computing laboratory through which we explore a range of issues concerning the use of media in the way we communicate and undertake creative endeavours. Visualisation in particular - data, program, knowledge - features in all research groupings. We pride ourselves on presenting a seamless view of computational paradigms and treat all such paradigms (symbolic as well as sub-symbolic, parallel as well as sequential, declarative as well as procedural, object-oriented as well as block-structured) as equally valid, according to the circumstances. With the formation of the new School, we aim to enhance joint research initiatives. Initially these might be on the basis of engineering applications providing a demonstrator for software technologies developed in computer science. Also the theoretical backing provided by research in computer science could aid developments in engineering. There will also be possibilities for technology led developments in media computing. It is the responsibility of the Director of each research group to produce an annual research plan based on the individual plans of the staff within the group. Such plans are the basis for recommendations on equipment purchase and access to travel funds and study leave. There are currently sixteen students on our MPhil/PhD programme. Since 1989 we have had a 100% submission record for all SERC/EPSRC funded PhD students. We run a thesis committee system whereby each student is encouraged in the development of their research but also has their progress monitored. We have an EPSRC-approved advanced, specialist MSc degree programme in New Generation Computing. Undergraduate Teaching We received a rating of excellent in 1994 HEFCE Teaching Quality Assessment, one of only 10 UK Universities so rated. Our submission to HEFCE stressed the importance of feeding research back into the teaching programmes and we apply this approach to all our degrees. Approximately 65 undergraduates are admitted each year. Our main programme is the Single Honours degree in Computer Science. In collaboration with the School of Psychology we offer an Interdisciplinary Single Honours programme in Cognitive Science. There are also in place arrangements for major/minor programmes with the School of Mathematical Sciences and a with European study programme. We have placed some of our modules within the University's modularity scheme thus allowing students from other disciplines to study computing. Teaching allocations are based on a formula that takes account of research supervisory commitments. Also, all junior staff are given time to undertake a structured independent-learning course (recognised by the Staff and Educational Development Association (SEDA)) run by the Staff Development Unit. Other Information The starting salary will be up to £26,523 on the in the Lecturer A scale or Lecturer B scale (£16,655 - £29,048), dependent upon teaching research experience. There is a 3-year probationary period. Full-time members of the academic staff may not undertake without consent of the University Council outside paid work, except external examining, literary work, public lecturing, and extra-mural teaching, occasional broadcasting and the giving of advice on an ad-hoc basis and on a relatively small scale. The successful applicant (if aged under 60) may join the Universities' Superannuation Scheme (USS), or maintain USS membership if applicable; 6.35% of gross salary is contributable by the individual, with the University contributing such sums as will be required to maintain the full benefits of the scheme. For permanent appointments, in the case of a person appointed from within the United Kingdom, all agreed removal expenses will normally be reimbursed. In the case of the appointment of a candidate from overseas, the full cost of removal expenses together with one adult economy fare from the candidate's place of residence will normally be reimbursed. In this context, it should be noted that the phrase "removal expenses" is interpreted by the University to mean the removal of household effects and excludes the cost of storage and the reimbursement of personal fares save the one adult economy fare payable to a person appointed overseas. Claims for reimbursement of removal expenses should be accompanied by three competitive tenders and the receipted account. Reimbursement will normally be on the basis of the lowest tender. Interviewees will be asked to complete a Medical Questionnaire. The University is committed to a comprehensive policy of equal opportunities in employment in which individuals are selected and treated solely on the basis of their relevant merits and abilities and are given equal opportunities within the University. In making this appointment the University reserves the right to consider persons other than those who have submitted applications by the closing date. To apply, candidates forward 7 copies (or, if posted overseas, 1 copy in a format suitable for photocopying) of your curriculum vitae, including particulars of your qualifications and experience, and a list of their publications, together with names and addresses of three referees a supporting letter of application which includes a statement indicating your intended areas of research, and summary of personal details sheet. Applications should reach the Personnel Division, University of Exeter, Northcote House, The Queen's Drive, Exeter, EX4 4QJ, Devon, UK, by 26 February 1999 . Please quote the reference number 6084 on your application and in all correspondence. Candidates should note that the University normally contacts referees for shortlisted candidates without further advice to applicants. When advising of referees, if they are resident overseas please ensure you supply fax numbers and/or email addresses enabling us to contact them (if necessary) without delay. In the interests of economy, the University does not notify candidates who have been unsuccessful with their application; consequently if you have not heard from us within six weeks of the closing date, please assume that you have not been successful on this occasion. For further information and an informal discussion please contact Dr Wendy Milne (w.milne at exeter.ac.uk), telephone 01392 264061. Informal contact with the Directors of research is also welcomed. fps.6084 From l.s.smith at cs.stir.ac.uk Fri Feb 19 08:28:06 1999 From: l.s.smith at cs.stir.ac.uk (Dr L S Smith (Staff)) Date: Fri, 19 Feb 99 13:28:06 GMT Subject: deadline extended for ICANN 99 - Ninth International Conference on Artificial Neural Networks Message-ID: <199902191328.NAA20500@tinker.cs.stir.ac.uk> ICANN 99 Submission Deadline - Extended ICANN 99, the annual conference of the European Neural Network Society held in conjunction with the IEE conference on Artificial Neural Networks, is taking place in Edinburgh, UK, from 07-10 September 1999. Due to numerous requests for bringing the ICANN 99 submission procedure into alignment with earlier years, the deadline for submissions has been revised to Friday 12 March 1999. For more information see http://www.iee.org.uk/Conf/ICANN/Welcome.html and for post-conference workshops see http://www.dai.ed.ac.uk/daidb/people/homes/ckiw/icann/ Leslie Smith Alan Murray David Willshaw For the Organising Committee From harnad at coglit.soton.ac.uk Sun Feb 21 08:04:58 1999 From: harnad at coglit.soton.ac.uk (Stevan Harnad) Date: Sun, 21 Feb 1999 13:04:58 +0000 (GMT) Subject: Law of Effect: BBS Call for Commentators Message-ID: Below is the abstract of a forthcoming BBS target article *** please see also 5 important announcements about new BBS policies and address change at the bottom of this message) *** BEHAVIORAL MOMENTUM AND THE LAW OF EFFECT by John A. Nevin & Randolph C. Grace This article has been accepted for publication in Behavioral and Brain Sciences (BBS), an international, interdisciplinary journal providing Open Peer Commentary on important and controversial current research in the biobehavioral and cognitive sciences. Commentators must be BBS Associates or nominated by a BBS Associate. To be considered as a commentator for this article, to suggest other appropriate commentators, or for information about how to become a BBS Associate, please send EMAIL before March 26 to: bbs at cogsci.soton.ac.uk or write to [PLEASE NOTE SLIGHTLY CHANGED ADDRESS]: Behavioral and Brain Sciences ECS: New Zepler Building University of Southampton Highfield, Southampton SO17 1BJ UNITED KINGDOM http://www.princeton.edu/~harnad/bbs/ http://www.cogsci.soton.ac.uk/bbs/ ftp://ftp.princeton.edu/pub/harnad/BBS/ ftp://ftp.cogsci.soton.ac.uk/pub/bbs/ If you are not a BBS Associate, please send your CV and the name of a BBS Associate (there are currently over 10,000 worldwide) who is familiar with your work. All past BBS authors, referees and commentators are eligible to become BBS Associates. To help us put together a balanced list of commentators, please give some indication of the aspects of the topic on which you would bring your areas of expertise to bear if you were selected as a commentator. An electronic draft of the full text is available for inspection with a WWW browser, anonymous ftp or gopher according to the instructions that follow after the abstract. _____________________________________________________________ BEHAVIORAL MOMENTUM AND THE LAW OF EFFECT John A. Nevin University of New Hampshire RR2, Box 162, Vineyard Haven, MA 02568 USA tnevin at worldnet.att.net Randolph C. Grace, University of Canterbury Department of Psychology, Christchurch, New Zealand r.grace at psyc.canterbury.ac.nz KEYWORDS: addiction, choice, extinction, generalization, law of effects, learning, momentum, movement, operant, Pavlov, preference, reinforcement, self-control, Thorndike ABSTRACT: In the metaphor of behavioral momentum, the rate of a free operant in the presence of a discriminative stimulus is analogous to the velocity of a moving body, and resistance to change measures an aspect of behavior that is analogous to its inertial mass. An extension of the metaphor suggests that preference measures an analog to the gravitational mass of that body. The independent functions relating resistance to change and preference to the conditions of reinforcement may be construed as convergent measures of a single construct, analogous to physical mass, that represents the effects of a history of exposure to the signaled conditions of reinforcement and that unifies the traditionally separate notions of the strength of learning and the value of incentives. Research guided by the momentum metaphor emcompasses the effects of reinforcement on response rate, resistance to change, and preference, and has implications for clinical interventions, drug addiction, and self-control. In addition, its principles can be seen as a modern, quantitative version of Thorndike's (1911) Law of Effect, providing a new perspective on some of the challenges to his postulation of strengthening by reinforcement. ____________________________________________________________ To help you decide whether you would be an appropriate commentator for this article, an electronic draft is retrievable from the World Wide Web or by anonymous ftp from the US or UK BBS Archive. Ftp instructions follow below. Please do not prepare a commentary on this draft. Just let us know, after having inspected it, what relevant expertise you feel you would bring to bear on what aspect of the article. The URLs you can use to get to the BBS Archive: http://www.princeton.edu/~harnad/bbs/ http://www.cogsci.soton.ac.uk/bbs/Archive/bbs.nevin.html ftp://ftp.princeton.edu/pub/harnad/BBS/bbs.nevin ftp://ftp.cogsci.soton.ac.uk/pub/bbs/Archive/bbs.nevin To retrieve a file by ftp from an Internet site, type either: ftp ftp.princeton.edu or ftp 128.112.128.1 When you are asked for your login, type: anonymous Enter password as queried (your password is your actual userid: yourlogin at yourhost.whatever.whatever - be sure to include the "@") cd /pub/harnad/BBS To show the available files, type: ls Next, retrieve the file you want with (for example): get bbs.nevin When you have the file(s) you want, type: quit ____________________________________________________________ *** FIVE IMPORTANT ANNOUNCEMENTS *** ------------------------------------------------------------------ (1) There have been some very important developments in the area of Web archiving of scientific papers very recently. Please see: Science: http://www.cogsci.soton.ac.uk/~harnad/science.html Nature: http://www.cogsci.soton.ac.uk/~harnad/nature.html American Scientist: http://www.cogsci.soton.ac.uk/~harnad/amlet.html Chronicle of Higher Education: http://www.chronicle.com/free/v45/i04/04a02901.htm --------------------------------------------------------------------- (2) All authors in the biobehavioral and cognitive sciences are strongly encouraged to archive all their papers (on their Home-Servers as well as) on CogPrints: http://cogprints.soton.ac.uk/ It is extremely simple to do so and will make all of our papers available to all of us everywhere at no cost to anyone. [Note, this is not addressed particularly to BBS authors, but to ALL authors of ALL articles in the biobehavioral and cognitive and related sciences.] --------------------------------------------------------------------- (3) BBS has a new policy of accepting submissions electronically. Authors can specify whether they would like their submissions archived publicly during refereeing in the BBS under-refereeing Archive, or in a referees-only, non-public archive. Upon acceptance, preprints of final drafts are moved to the public BBS Archive: ftp://ftp.princeton.edu/pub/harnad/BBS/.WWW/index.html http://www.cogsci.soton.ac.uk/bbs/Archive/ -------------------------------------------------------------------- (4) BBS has expanded its annual page quota and is now appearing bimonthly, so the service of Open Peer Commentary can now be be offered to more target articles. The BBS refereeing procedure is also going to be considerably faster with the new electronic submission and processing procedures. Authors are invited to submit papers to: Email: bbs at cogsci.soton.ac.uk Web: http://cogprints.soton.ac.uk http://bbs.cogsci.soton.ac.uk/ INSTRUCTIONS FOR AUTHORS: http://www.princeton.edu/~harnad/bbs/instructions.for.authors.html http://www.cogsci.soton.ac.uk/bbs/instructions.for.authors.html --------------------------------------------------------------------- (5) Call for Book Nominations for BBS Multiple Book Review In the past, Behavioral and Brain Sciences (BBS) journal had only been able to do 1-2 BBS multiple book treatments per year, because of our limited annual page quota. BBS's new expanded page quota will make it possible for us to increase the number of books we treat per year, so this is an excellent time for BBS Associates and biobehavioral/cognitive scientists in general to nominate books you would like to see accorded BBS multiple book review. (Authors may self-nominate, but books can only be selected on the basis of multiple nominations.) It would be very helpful if you indicated in what way a BBS Multiple Book Review of the book(s) you nominate would be useful to the field (and of course a rich list of potential reviewers would be the best evidence of its potential impact!). From mw at stat.Duke.EDU Mon Feb 22 11:18:26 1999 From: mw at stat.Duke.EDU (Mike West) Date: Mon, 22 Feb 1999 11:18:26 -0500 Subject: 1999 MITCHELL PRIZE: revisited Message-ID: <19990222111826.A14866@isds.duke.edu> This is a reposting of the 1999 Mitchell Prize announcement, correcting a typo and clarifying eligibility. With apologies for multiple postings, we'd appreciate your help in bringing this to the attention of potential nominators and future Mitchell Prize winners! Thanks for your help. ---------------------------------------------------------------------- 1999 MITCHELL PRIZE: ANNOUNCEMENT AND SOLICITATION http://www.stat.duke.edu/sites/mitchell.html The third Mitchell Prize will be awarded in 1999. The Prize will be announced and presented at the August 1999 Joint Statistical Meetings in Baltimore. The Prize is awarded in recognition of an outstanding paper describing how a Bayesian analysis has solved an important applied problem. The 1999 Prize consists of an award of $1000 and a commemorative plaque. The Mitchell Prize is named for Toby J. Mitchell and was established by his friends and colleagues following his death from leukemia in 1993. Toby was a Senior Research Staff Member at Oak Ridge National Laboratory throughout his career, will leaves of absence spent at the University of Wisconsin and at the National Institute of Environmental Health Sciences. Toby won the Snedecor Award in 1978 (with co-author Bruce Turnbull), made incisive contributions to statistics, especially in biometry and engineering applications, and was a marvelous collaborator and an especially thoughtful scientist. Toby was a dedicated Bayesian, hence the focus of the prize. This is the third Mitchell Prize, the first two having been awarded in 1994 and 1997. Beginning this year, the Prize will be awarded annually under the cosponsorship of the ASA Section on Bayesian Statistical Science (SBSS), the International Society for Bayesian Analysis (ISBA), and the Mitchell Prize Founders' Committee. The sponsors annually establish a selection committee; the 1999 Prize selection committee consists of Gary Koop, Max Morris (chair) and Mike West. To be eligible for the 1999 Prize, a paper will either have appeared in a refereed journal or refereed conference proceedings since January 1 1997, or be scheduled for future publication in a refereed outlet. Candidate papers will be accepted from nominators and from authors. In reviewing submissions, emphasis will be placed on evidence that the application has truly benefited from a Bayesian analysis respecting the individual character of the problem at hand. There is no restriction as to approach taken, except that it be Bayesian in some sense, and that it carefully and appropriately justifies models, priors and methodologies adopted. To be considered for this year's Mitchell Prize, please send four reprints or copies of the manuscript along with a cover letter that includes: a brief statement of the impact of the work, and contact information for the authors, nominator (if not an author), and two individuals who are not statisticians or coauthors and who can be contacted for an evaluation of the importance of the work in the applied field. Submissions should be mailed to Max Morris Department of Statistics Iowa State University Snedecor Hall Ames, IA 50011-1210 Entries must be received by May 15, 1999 in order to receive consideration. For further information, phone 515-294-2775, FAX 515-294-4040, or e-mail mmorris at iastate.edu. To learn more about the sponsoring organizations SBSS and ISBA visit their web sites at www.stat.duke.edu/sbss/sbss.html and www.bayesian.org/ respectively. ------------------------------------------------------------------------- From istvan at usl.edu Mon Feb 22 17:48:47 1999 From: istvan at usl.edu (Dr. Istvn S. N. Berkeley) Date: Mon, 22 Feb 1999 16:48:47 -0600 Subject: Cog. Sci. Job Message-ID: <36D1DECF.37D@USL.edu> The University of Southwestern Louisiana's Institute for Cognitive Science invites applications for a tenure-track faculty appointment at the assistant professor level for a new Ph.D. program. Appointment is anticipated for the Fall of 1999. Applicants must have an earned doctorate in cognitive science or a closely related area, as well as demonstrated potential for research excellence. Focus areas of the new program are in cognitive processes, comparative cognition, computational models of mind, cognitive neuroscience, and language and thought. Salary is competitive, depending on experience and qualifications. Applicants should send a letter of application, statement of research interests, curriculum vitae, reprints, and three letters of reference to Dr. Daniel Povinelli, Institute for Cognitive Science, P.O. Drawer 43772, University of Southwestern Louisiana, Lafayette, Louisiana 70504. Review of applications will commence March 15, 1999. The University is in compliance with Title IX of the Civil Rights Act, Section 504 of the Rehabilitation Act of 1973, and is an Equal Employment Opportunity Affirmative Action Employer. -- Istvan S. N. Berkeley Ph.D, E-mail: istvan at USL.edu, Philosophy, The University of Southwestern Louisiana, USL P. O. Box 43770, Lafayette, LA 70504-3770, USA. Tel:(318) 482 6807, Fax: (318) 482 6195, http://www.ucs.usl.edu/~isb9112 From amari at brain.riken.go.jp Tue Feb 23 05:27:25 1999 From: amari at brain.riken.go.jp (Shunichi Amari) Date: Tue, 23 Feb 1999 19:27:25 +0900 Subject: Brain Science Summer Course Message-ID: <19990223192725G.amari@brain.riken.go.jp> The RIKEN Brain Science Institute is offering a summer program to train advanced students interested in molecular, cellular, systems, or computational neuroscience. Applicants may choose either a laboratory internshipfor up to two months, or participate in an intensive two-week lecture course featuring a distinguished international faculty. See our website for details. Summer Internship: July - September, 1999 Lecture Course: July 25 - August 7, 1999 http://www.brain.riken.go.jp/summer/bsi-summer.html Shun-ichi Amari From jose.m.bernardo at uv.es Tue Feb 23 06:41:16 1999 From: jose.m.bernardo at uv.es (Jose M. Bernardo) Date: Tue, 23 Feb 1999 12:41:16 +0100 Subject: Workshop of Bayesian Objective Methodology Message-ID: Apologies for cross-posting INTERNATIONAL WORKSHOP ON OBJECTIVE BAYESIAN METHODOLOGY SECOND INFORMATION BULLETIN 1499--1999: Five Centuries of the Universitat de Val?ncia Valencia, Spain, June 11th -- June 13th, 1999 The Universitat of Val?ncia, Spain, will celebrate next year its 500 anniversary. Within the Centenary Programme, the University of Valencia is sponsoring an International Workshop on Objective Bayesian Methodology Objective Bayesian Methodology: Objective Bayesian methodology is of increasing importance today for at least two reasons. First, application of Bayesian analysis is rapidly growing among nonspecialists, most of whom seek automatic or objective Bayesian procedures. Second, computational advances have allowed Bayesian methodology to be employed in problems of such complexity that determination of serious subjective priors is essentially impossible. The corresponding explosion of research on objective Bayesian methods makes a workshop in this area particularly timely. Objective Bayesian methodology is, for the most part, oriented towards the development of prior distributions that can be used automatically, ie that do not require subjective input. There are three quite distinct statistical domains in which this development has taken place: parametric estimation, parametric testing, and nonparametric analysis. The specific topics in these areas that are currently being considered for the Workshop are listed below. Note, however, that it is planned to maintain considerable flexibility so as to be able to include late-breaking developments. 1. Objective Priors and Frequentist Statistics 2. Determination of Objective Priors in Important Problems 3. Priors for Objective Bayesian Testing and Model Selection 4. Objective Priors in Nonparametric Analysis 5. The Roles of Objective Bayesian Analysis Workshop Organization: The three day workshop will take place at the Peset Center, a reconverted XVIII century palace in the medieval old town of the city of Valencia, with appropriate facilities. Workshop Web sites: http://www.uv.es/~bernardo/workshop.html (Spain) http://www.stat.duke.edu/~bernardo/workshop.html (USA mirror) Programme Committee: Professor James O. Berger ({ Duke University, USA}) Professor Jos? M. Bernardo ({ Universitat de Val?ncia, Spain}) Professor Jayanta K. Ghosh ({ Indian Statistical Institute, India and Purdue University, USA}) Professor Malay Ghosh ({ University of Florida, USA}) Professor Christian P. Robert ({ Universit? de Rouen, France}) Scientific Programme: The scientific programme will feature the following invited lectures Berger, James, Duke University, USA (Generalized) Reference Priors for Spatial Models (with B. Sanso, and V. De Oliveira) Bernardo, Jos?, Universidad de Valencia, Spain Bayesian Reference Multivariate Hypothesis Testing (with Rueda) Datta, Gauri, University of Georgia, USA Bayesian Prediction with Approximate Frequentist Validity Dawid, a. Phil, University College London, UK The Trouble with Bayes Factors Fraser, Donald, University of Toronto, Canada Objective priors and model dependence: the frequentist-Bayesian interface Ghosh, Jayanta, Indian Statistical Institute, India Prior Elicitation, Default Priors and Prediction Loss Ghosh, Malay, University of Florida, USA Noninformative Priors for Item Response Models Ghosal, Subhashis, Vrije Universiteit Amsterdam, the Netherlands Default Priors, Model Selection and Adaptation in Density Estimation Liseo, Brunero, Universit? di Roma La Sapienza, Italy, Semiparametric Inference for Long Memory Time Series: on Bayesian-Frequentist Interface (with D. Marinucci and L. Petrella) Mukerjee, Rahul, Indian Institute of Management, India On Confidence Intervals Associated with the Usual and Adjusted Likelihoods (with N. Reid) P?rez, Jos? M., Universidad Simon Bolivar, Venezuela Expected Posterior Priors for Default Model Selection Pericchi, Luis, Universidad Sim?n Bol?var, Venezuela The Intrinsic Bayes Factor as a Constructive Method to Develop Default or Conventional Priors for Hypothesis Testing and Model Selection. Robert, Christian, CREST-ENSAE, France Nonidentifiable Models and Symmetric Priors: How to Teach MCMC Samplers to Jump Farther (with M. Hurn) Rousseau, Judith, CREST-ENSAE, France Edgeworth expansion for a Gaussian long-memory process with application to matching priors Sun, Dongchu, University of Missouri-Columbia, USA Random Effects in Generalized Linear Mixed Models Sweeting, Trevor, University of Surrey, UK Coverage Probability Bias, Objective Bayes and the Likelihood Principle Wasserman, Larry, Carnegie Mellon University, USA Priors for Infinite Dimensional Problems Each lecture will be followed by invited discussions and by discussion from the floor. Invited discussants include: Barbieri, Marilena, Universit? di Roma La Sapienza, Italy Bayarri, Susie, Universitat de Val?ncia, Spain Clarke, Bertrand, University of British Columbia, Canada Clyde, Merlise, Duke University , USA de Santis, Fulvio, Universit? di Roma La Sapienza, Italy DiCiccio, Thomas, Stanford University, USA Gelfand, Alan, University of Connecticut, USA George, Edward, University of Texas, USA Gir?n, Javier, Universidad de M?laga, Spain Guti?rrez-Pe?a, Eduardo, UNAM, Mexico Mendoza, Manuel, ITAM, Mexico Moreno, El?as, Universidad de Granada, Spain Mortera, Julia, Universit? di Roma 3, Italy Petrone, Sonia, Universit? di Pavia, Italy Ramamoorthi, R., Michigan State University, USA Reid, Nancy, University of Toronto, Canada Rueda, Ra?l, UNAM, Mexico Severini, Thomas, Northwestern University, USA Sivaganesan, Siva, University of Cincinnati, USA van der Linde, Angelika, University of Edinburgh, UK Wells, Martin, Cornell University, USA Ye, Keying, Virginia Polytechnic Institute and State University, USA Contributed papers are encouraged. These will be presented in two plenary poster sessions, organized in the tradition of the Valencia meetings. The abstracts deadline is April 30th . The abstract should be sent by e-mail using the abstract format http://www.uv.es/~bernardo/abstract.html (Spain) http://www.stat.duke.edu/~bernardo/abstract.html (USA mirror) accessible at the workshop web sites. There will not be published proceedings, but a booklet of extended abstracts will be distributed at the Workshop. Travel Information: Delegates will be expected to arrive on Thursday June 10th evening, and to depart on Monday 14th morning. The opening ceremony will be on the morning of Friday 11th, and the Conference Dinner on the evening of Sunday 13th. Valencia international airport, 7 km. from town, has direct connections to many european cities, and frequent connexions to both Madrid and Barcelona. Registration: To register, please mail or fax the registration form http://www.uv.es/~bernardo/registration.html (Spain) http://www.stat.duke.edu/~bernardo/registration.html (USA mirror) accessible at the workshop web sites. Those possibly interested in participating, but not ready to register, should e-mail the local organizer, Jos? M. Bernardo mailto:jose.m.bernardo at uv.es Periodic electronic Information Bulletins, will be sent to those who who have declared their interest. Additional information: Further information on the workshop, including pointers to the first draft of the talks as they become available is posted in the workshop web sites. http://www.uv.es/~bernardo/workshop.html (Spain) http://www.stat.duke.edu/~bernardo/workshop.html (USA mirror) For tourist information on the area, please visit he Web site of the Tourism Agency of the State of Valencia, http://www.comunitat-valenciana.com PLEASE NOTE: On April 4th, all Spanish telephone numbers added a "9" after the country code "34'. Thus, my more direct (phone and fax) line is now: +34.96.364.3560 ____________________________________________________________________ Prof. Dr. Jose M. Bernardo, Universidad de Valencia Dept. Estadistica, Fac. Matematicas, 46100-Burjassot, Valencia, Spain Tel. +34.96.364.3560 and +34.96.386.4313 (direct), +34.96.386.4362 (office). Fax: +34.96.364.3560 (direct), +34.96.386.4735 (office). E Mail: jose.m.bernardo at uv.es, Web Pages: http://www.uv.es/~bernardo/ or http://www.stat.duke.edu/~bernardo/ ____________________________________________________________________ From Jerry.Friedman at cmis.CSIRO.AU Tue Feb 23 21:12:34 1999 From: Jerry.Friedman at cmis.CSIRO.AU (Jerry.Friedman@cmis.CSIRO.AU) Date: Wed, 24 Feb 1999 13:12:34 +1100 Subject: Boosting methods for regression and classification. Message-ID: <199902240212.NAA18316@pride.nsw.cmis.CSIRO.AU> *** Technical Report Available *** Greedy Function Approximation: A Gradient Boosting Machine Jerome H. Friedman Stanford University ABSTRACT Function approximation is viewed from the perspective of numerical optimization in function space, rather than parameter space. A connection is made between stagewise additive expansions and steepest-descent minimization. A general gradient-descent "boosting" paradigm is developed for additive expansions based on any fitting criterion. Specific algorithms are presented for least-squares, least-absolute-deviation, and Huber-M loss functions for regression, and multi-class logistic likelihood for classification. Special enhancements are derived for the particular case where the individual additive components are decision trees, and tools for interpreting such "TreeBoost" models are presented. Gradient boosting of decision trees produces competitive, highly robust, interpretable procedures for regression and classification, especially appropriate for mining less than clean data. Connections between this approach and the boosting methods of Freund and Shapire 1996, and Friedman, Hastie, and Tibshirani 1998 are discussed. Available from: "http://www-stat.stanford.edu/~jhf/ftp/trebst.ps" From Jon.Baxter at syseng.anu.edu.au Wed Feb 24 08:14:50 1999 From: Jon.Baxter at syseng.anu.edu.au (Jonathan Baxter) Date: Thu, 25 Feb 1999 00:14:50 +1100 (EST) Subject: Boosting methods for regression and classification. In-Reply-To: <199902240212.NAA18316@pride.nsw.cmis.CSIRO.AU> from "Jerry.Friedman@cmis.CSIRO.AU" at Feb 24, 99 01:12:34 pm Message-ID: <199902241314.AAA20350@reid.anu.edu.au> > > > > *** Technical Report Available *** > > > Greedy Function Approximation: > A Gradient Boosting Machine > > Jerome H. Friedman > Stanford University > > ABSTRACT > > Function approximation is viewed from the perspective of numerical > optimization in function space, rather than parameter space. A > connection is made between stagewise additive expansions and > steepest-descent minimization. A general gradient-descent "boosting" > paradigm is developed for additive expansions based on any fitting > criterion. > Specific algorithms are presented for least-squares, > least-absolute-deviation, and Huber-M loss functions for regression, > and multi-class logistic likelihood for classification. Special > enhancements are derived for the particular case where the individual > additive components are decision trees, and tools for interpreting > such "TreeBoost" models are presented. Gradient boosting of decision > trees produces competitive, highly robust, interpretable procedures > for regression and classification, especially appropriate for mining > less than clean data. Connections between this approach and the > boosting methods of Freund and Shapire 1996, and Friedman, Hastie, and > Tibshirani 1998 are discussed. > > > Available from: "http://www-stat.stanford.edu/~jhf/ftp/trebst.ps" There was also some discussion of the connection between boosting and gradient descent in function space at the NIPS workshop on large margins in December. I have put the slides of my talk on the subject---"AnyBoost: Boosting with (almost) arbitrary cost functions and steps"---on my web page for those who are interested (http://syseng.anu.edu.au/~jon/anyboost.ps). Cheers, Jon ------------- Jonathan Baxter Research Fellow Department of Systems Engineering Research School of Information Science and Engineering Australian National University http://syseng.anu.edu.au/~jon Tel: +61 2 6279 8678 Fax: +61 2 6279 8688 > > > From rsun at research.nj.nec.com Wed Feb 24 13:43:40 1999 From: rsun at research.nj.nec.com (Ron Sun) Date: Wed, 24 Feb 1999 13:43:40 -0500 Subject: TR on reinforcement learning Message-ID: <199902241843.NAA00660@pc-rsun.nj.nec.com> Announcing three new papers on combining reinforcement learning with symbolic methods: --------------------------------- Autonomous Learning of Sequential Tasks: Experiments and Analyses by Ron Sun$^{1,2}$, Todd Peterson$^2$ $^1$ NEC Research Institute, 4 Independence Way, Princeton, NJ 08540 \\ $^2$ The University of Alabama, Appeared in IEEE Transactions on Neural Networks, Nov.1998 http://cs.ua.edu/~rsun/sun.tnn98.ps ABSTRACT: This paper presents a novel learning model {\sc Clarion}, which is a hybrid model based on the two-level approach proposed in Sun (1995). The model integrates neural, reinforcement, and symbolic learning methods to perform on-line, bottom-up learning (i.e., learning that goes from neural to symbolic representations). The model utilizes both procedural and declarative knowledge (in neural and symbolic representations respectively), tapping into the synergy of the two types of processes. It was applied to deal with sequential decision tasks. Experiments and analyses in various ways are reported that shed light on the advantages of the model. --------------------------------- Multi-Agent Reinforcement Learning: Weighting and Partitioning by Ron Sun and Todd Peterson To appear in: Neural Networks http://cs.ua.edu/~rsun/sun.NN99.ps ABSTRACT: This paper addresses weighting and partitioning in complex reinforcement learning tasks, with the aim of facilitating learning. The paper presents some ideas regarding weighting of multiple agents and extends them into partitioning an input/state space into multiple regions with differential weighting in these regions, to exploit the differential characteristics of regions and the differential characteristics of agents to reduce the learning complexity of agents (and their function approximators) and thus to facilitate the learning overall. It analyzes, in reinforcement learning tasks, different ways of partitioning a task and using agents selectively based on partitioning. Based on the analysis, some heuristic methods are described and experimentally tested. We find that some off-line heuristic methods performed the best, significantly better than single-agent models. --------------------------------- A Hybrid Architecture for Situated Learning of Reactive Sequential Decision Making by Ron Sun Todd Peterson Edward Merrill To appear in: Applied Intelligence http://cs.ua.edu/~rsun/sun.apin99.ps ABSTRACT: In developing autonomous agents, one usually emphasizes only (situated) procedural knowledge, ignoring more explicit declarative knowledge. On the other hand, in developing symbolic reasoning models, one usually emphasizes only declarative knowledge, ignoring procedural knowledge. In contrast, we have developed a learning model {\sc Clarion}, which is a hybrid connectionist model consisting of both localist and distributed representations, based on the two-level approach proposed in Sun (1995). {\sc Clarion} learns and utilizes both procedural and declarative knowledge, tapping into the synergy of the two types of processes, and enables an agent to learn in situated contexts and generalize resulting knowledge to different scenarios. It unifies connectionist, reinforcement, and symbolic learning in a synergistic way, to perform on-line, bottom-up learning. This summary paper presents one version of the architecture and some results of the experiments. ----------------------------------------- Dr. Ron Sun NEC Research Institute 4 Independence Way Princeton, NJ 08540 phone: 609-520-1550 fax: 609-951-2483 email: rsun at cs.ua.edu, rsun at research.nj.nec.com (July 1st, 1998 -- July 1st, 1999) ----------------------------------------- Prof. Ron Sun http://cs.ua.edu/~rsun Department of Computer Science and Department of Psychology phone: (205) 348-6363 The University of Alabama fax: (205) 348-0219 Tuscaloosa, AL 35487 email: rsun at cs.ua.edu From B.Graham at anc.ed.ac.uk Thu Feb 25 12:59:03 1999 From: B.Graham at anc.ed.ac.uk (Bruce Graham) Date: Thu, 25 Feb 1999 17:59:03 GMT Subject: ICANN*99 Call for Tutorial Proposals Message-ID: <199902251759.RAA01257@dingo.cns.ed.ac.uk> CALL FOR PROPOSALS ICANN*99 Conference Tutorials 7 September 1999 Edinburgh, Scotland Tutorials on current topics in the fields of research covered by ICANN99 will be held on Tuesday 7 September 1999, the first day of the ICANN conference program. Proposals by qualified individuals interested in giving a half-day tutorial are solicited. Topics should fall within the broadly defined area of neural networks research, including: Theory and Algorithms; Neurobiology and Computational Neuroscience; Cognitive Modelling; Industrial, Commercial and Medical Applications; Hardware and Neuromorphic Engineering; Control, Robotics and Adaptive Behaviour. It is planned to run at least 6 tutorials of around 3 hours each: 3 in parallel in the morning and another 3 in the afternoon. Attendees will pay a registration fee and tutors will be given some recompense for their time and effort. Consequently we seek very high quality proposals. This call for proposals can be found at: http://www.iee.org.uk/Conf/ICANN/ and further information will be added as it becomes available. Submission Instructions ----------------------- Interested parties should submit via e-mail a short proposal for a tutorial of interest by 29 March, 1999 to Dr Bruce Graham (B.Graham at ed.ac.uk), Institute for Adaptive and Neural Computation, University of Edinburgh. Proposals should include an outline of the tutorial contents and your background and qualifications for giving such a tutorial. The final programme of tutorials will be decided by the middle of April. If e-mail is unavailable, mail or fax so as to arrive by the deadline to: ICANN*99 Tutorials c/o Dr Bruce Graham Institute for Adaptive and Neural Computation Division of Informatics, University of Edinburgh 2 Buccleuch Place, Edinburgh EH8 9LW, Scotland, UK Fax: +44-131-650-4406 PROPOSALS MUST BE RECEIVED BY MARCH 29 1999 (Bruce Graham, IANC, Division of Informatics, University of Edinburgh) From torras at iri.upc.es Thu Feb 25 07:29:43 1999 From: torras at iri.upc.es (Carme Torras) Date: Thu, 25 Feb 1999 13:29:43 +0100 (MET) Subject: Reminder: CFP Adaptive Robots Special Issue Message-ID: <199902251229.NAA05286@sibelius.upc.es> This is being mailed to several mailing lists. Please accept our apologies if you receive multiple copies. =============================================================================== Special Issue of the journal CONNECTION SCIENCE on *** ADAPTIVE ROBOTS *** CALL FOR PAPERS: Deadline March 15th, 1999 Adaptivity is the capability of self-modification that some agents have, which allows them to maintain a level of performance when facing environmental changes, or to improve it when confronted repeatedly with the same situation. This special issue is aimed at capturing the state of the art in the intricate task of endowing robots with adaptive capabilities, with a special emphasis on neural-based solutions. Thus, some examples of topics covered are: - Adaptive sensing - Adaptive gaits for walking robots - Self-calibration of robot manipulators - Adaptive dynamic control of flexible robot arms - Acquiring fine manipulation skills - Learning hand-eye coordination - Exploration and reinforcement learning - Improving robot navigation - Adaptive multi-robot systems The special issue will adhere to an engineering perspective, i.e. the emphasis will be on solving practical robotic problems using adaptive techniques, disregarding their possible biological (or cognitive) inspiration or plausibility. Work on real robots is preferred, with special attention being devoted to replicability of results, as well as to the discussion of the limitations (together with the advantages) of the proposed techniques. Guest editor: ------------- Carme TORRAS, CSIC-UPC (Spain) Editorial board: ---------------- Rudiger DILLMANN, University of Karlsruhe (Germany) Leslie P. KAELBLING, Brown University (USA) Ben KROSE, University of Amsterdam (The Netherlands) Jose R. MILLAN, Joint Research Centre (Italy) Helge RITTER, University of Bielefeld (Germany) Shankar SASTRY, University of California at Berkeley (USA) Noel SHARKEY, University of Sheffield (UK) Tim SMITHERS, CEIT (Spain) Tom ZIEMKE, University of Skovde (Sweden) Submissions to this special issue should be sent by March 15th, 1999 to: Carme Torras Institut de Robotica i Informatica Industrial (CSIC-UPC) Gran Capita 2-4 (edifici Nexus) 08034-Barcelona (Spain) e-mail: ctorras at iri.upc.es http://www-iri.upc.es/people/torras SCHEDULE: --------- March 15th, 99 - submission deadline May 15th, 99 - information to authors July 1st, 99 - deadline for final papers October 99 - publication of the special issue *** CONNECTION SCIENCE *** Journal of Neural Computing, Artificial Intelligence and Cognitive Research http://www.carfax.co.uk/cos-ad.htm =============================================================================== From A.van.Ooyen at nih.knaw.nl Fri Feb 26 11:24:01 1999 From: A.van.Ooyen at nih.knaw.nl (Arjen van Ooyen) Date: Fri, 26 Feb 1999 17:24:01 +0100 Subject: Poly-and Mononeuronal Innervation in a Model of Neuromuscular Junction Message-ID: <36D6CAA1.240E@nih.knaw.nl> Poly- and Mononeuronal Innervation in a Model for the Development of Neuromuscular Connections A. van Ooyen & D. J. Willshaw J. Theor. Biol. (1999) 196: 495-511. Download: http://www.cns.ed.ac.uk/people/arjen/papers/polyneuronal_net.ps.gz or request reprint: A.van.Ooyen at nih.knaw.nl http://www.cns.ed.ac.uk/people/arjen/competition.html ABSTRACT In the normal development of connections between motor neurons and muscle fibres, an initial stage of polyneuronal innervation is followed by withdrawal of connections until each muscle fibre is innervated by a single axon. However, polyneuronal innervation has been found to persist after prolonged nerve conduction block, in spite of the resumption of normal neuromuscular activity. Here we analyse in detail a model proposed for the withdrawal of nerve connections in developing muscle, based on competition between nerve terminals. The model combines competition for a presynaptic resource with competition for a postsynaptic resource. Using bifurcation and phase space analysis, we show that polyneuronal innervation, as well as mononeuronal innervation, can be stable. The model accounts for the development of mononeuronal innervation as well as for persistent polyneuronal innervation after prolonged nerve conduction block, which appears as a consequence of the general competitive interactions operating during normal development. -- Arjen van Ooyen, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. email: A.van.Ooyen at nih.knaw.nl website: http://www.cns.ed.ac.uk/people/arjen.html phone: +31.20.5665483 fax: +31.20.6961006 From dorota at at.net.au Sun Feb 28 08:11:19 1999 From: dorota at at.net.au (Dorota Kieronska) Date: Sun, 28 Feb 1999 21:11:19 +0800 Subject: CFP: ICONIP'99 Message-ID: <51540563F42FD2119B8600002139404504F3F3@tower0.at.net.au> ************************************** * * * ICONIP'99 * * jointly with * * ANZIIS'99 and ANNES'99 and ACNN'99 * * * ************************************** The 6th International Conference on Neural Information Processing-- jointly with the 7th Australian and New Zealand International Conference on Intelligent Information Processing Systems, and the 5th New Zealand International Conference on Artificial Neural Networks and Expert Systems, and the 11th Australian Conference on Neural Networks. ************************ * * * 16-20 November, 1999 * * Perth, Australia * * * ************************ In 1999, the annual conference of the Asian Pacific Neural Network Assembly, ICONIP'99, will be held jointly with three other major international conferences in the Asian Pacific Region, from 16 to 20 November 1999 in Perth, Western Australia. The joint conference will have three parallel streams: ************************************************************ * * Stream1: Biological Neural Information Processing * * Stream2: Neural Information Processing and Soft Computing * * Stream3: Applied Techniques * * ************************************************************ TOPICS OF INTEREST Stream 1: Biological Neural Information Processing Neurobiological systems Cognition Cognitive models of the brain Dynamical modelling, chaotic processes in the brain Brain computers, biological computers Consciousness, awareness, attention Adaptive biological systems Modelling emotions Perception, vision Learning languages Stream 2: Neural Information Processing and Soft Computing Soft computing--paradigms, methods, tools Artificial neural network models, architectures and algorithms Evolutionary programming and genetic algorithms Fuzzy systems Optimisation Expert systems Artificial life Adaptive systems Approximate reasoning Hybrid and ensemble systems Distributed AI systems, agent-based systems Machine learning, data mining and intelligent databases Probabilistic and statistical methods Stream 3: Applied techniques and Intelligent information systems Intelligent information retrieval systems Pattern recognition and image processing Speech recognition and language processing Human-computer interfaces, web computing Robotics and mechatronics Information engineering Time-series prediction Intelligent control Virtual reality Application of intelligent information technologies in: engineering, process industries, law, finance and business, medicine and sport, geographic information systems, geology, data mining, environmental protection, ... ************************************************************************ HONORARY CHAIR Shun-Ichi Amari, Tokyo University GENERAL CONFERENCE CHAIR, Tamas (Tom) Gedeon, University of New South Wales GENERAL CONFERENCE CO-CHAIR Yianni Attikiouzel, University of Western Australia Marwan Jabri, University of Sydney Nikola Kasabov, University of Otago PROGRAM COMMITEE CO-CHAIRS K Fukushima (jp), T Gedeon, S Halgamuge (au), N Kasabov (nz), D Nauck (de) INTERNATIONAL PROGRAM COMMITTEE (TENTATIVE) T Adams (fj), D Aha (us), D Alahakoon (au), I Aleksander (uk), S Amari (jp), J Andreae (nz), Y Attikiouzel (au), J Austin (uk), J Baldwin (jp), P Bartlett (au), A Bastian, J Bezdek (us), R Bogner (au), B Bouchon-Meunier (fr), B Bruegge (de), A Bulsara (us), T Caelli (au), T Chen (cn), V Cherkassky (us), S Cho, S Cho (kr), A Cichocki (jp), I Cloete (za), R Coggins (au), G Coghill (nz), G Deboeck (us), J Diederich, V Dimitrov (au), L Ding (sg), T Downs (au), W Duch (pl), W Dunin-Barkowski (su), R Eckmiller (de), F Esteva (es), E Falkenauer (be), M Fedrizzi (it), B Flower (au), D Fogel (us), W Friedrich (nz), K Fukushima, T Furuhashi (jp), T Gedeon (au), M Glesner (de), J Grantner (us), S Halgamuge (au), M Hassoun (us), H Hellendoorn (nl), T Hendtlass (au), K Hirota (jp), A Hoffmann, J Hughes (au), M Ishikawa (jp), M Jabri, I Jagielska, L Jain (au), M Jamshidi (us), A Jennings (au), S Jones (uk), J Kacprzyk (pl), V Karri (au), N Kasabov (nz), O Katai, H Kawakami (jp), O Kaynak (tr), L Koczy, S Kovacs (hu), A Kowalczyk (au), W Kurutach (th), R Kozma (nz), R Kruse (de), D Lakov (bg), S Lee (kr), B Lees, D Levy (au), M Lim (sg), P Maheshwari (au), R Marks (us), F Masulli (it), G Matsumoto (jp), D McMichael (au), J Miller (nz), S Mitra (us), M Mizumoto (jp), N Morgan (us), M Mukaidono, T Musha, T Nagano (jp), D Nauck (de), M Negnevitsky (au), M Negoita (us), C Newton, C Latimer (au), R O'Shea (nz), E Oja (fi), Y Okabe (jp), N Pal, S Pal (in), M Palaniswami (au), G Pasi (it), L Patnaik (in), M Paulin (nz), D Pham (uk), A Ralescu (tr), B Reusch (de), L Reznik (au), D Ruan (be), I Rudas (hu), P Sallis (nz), E Sanchez (fr), D Saffen, Y Sato (jp), B Schuermann (de), O Simula (fi), U Srinivasan (au), S Sugiyama (jp), R Sun, H Szu (us), M Takacs (yu), H Takagi (jp), J Taylor (uk), A Topchy (ru), P Treleavan (uk), N Tschiold-Guerman (ch), A Tsoi (au), Y Tsujimura, M Tsukada, E Uchino, S Usui (jp), J Herik (nl), V Vapnik, R Vemuri (us), N Vojdani, W Wahlster (de), A Waibel (us), P Wallis, J Wiles, W Wilson, P Wong (au), L Xu (hk), T Yamakawa, T Yanaru (jp), X Yao (au), D Yun, L Zadeh, J Zurada (us) ORGANISING COMMITTEE CO-CHAIRS: Patrick Wong (Chair), Tom Gedeon, Graham Mann, University of NSW CONFERENCE ORGANISER Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au CALL FOR PAPERS Papers must be received by 16 April 1999. They will be reviewed by senior researchers in the field and the authors will be informed about the decision of the review process in July 1999. The accepted papers must be submitted in a camera-ready format by 31 August. All accepted papers will be published by IEEE Computer Society Press. As the conference is a multi-disciplinary meeting the papers are required to be comprehensible to a wider rather than to a very specialised audience. Papers will be presented at the conference either in an oral or in a poster session. Please submit five copies of the paper written in English on A4-format white paper with 2.5 cm margins on all four sides, in two column format, on not more than 6 pages, single-spaced, in Times or similar font of 10 points, and printed on one side of the page only. Centred at the top of the first page should be the complete title, author(s), mailing and e-mailing addresses, followed by an abstract and the text. In the covering letter the stream and the topic of the paper according to the list above should be indicated. The IEEE Transaction journals LaTex article style can be used. SPECIAL ISSUES OF JOURNALS AND EDITED VOLUMES Selected papers will be published in special issues of scientific journals. The organising committee is looking for publications of edited volumes which include chapters covering the conference topics written by invited conference participants. TUTORIALS (16 November) Conference tutorials will be organized to introduce the basics of cognitive modelling, dynamical systems, neural networks, fuzzy systems, evolutionary programming, soft computing, expert systems, hybrid systems, and adaptive systems. Proposals for tutorials are due on 15 March 1999. EXHIBITION Companies and university research laboratories are encouraged to exhibit their developed or distributing software and hardware systems. STUDENT SESSION Postgraduate students are encouraged to submit papers to this session following the same formal requirements for paper submission. The submitted papers will be published in a separate brochure. VENUE (Perth) The Conference will be held in Perth city. The closing session will be held on Saturday, 20 November at a scenic bush location following an outdoor seafood barbecue. ACCOMMODATION Accommodation has been booked at the conference venue, and in the numerous Hotels nearby suitable for a range of budgets. Please see the web site for details after the paper deadline date. POSTCONFERENCE EVENTS The ICONIP'99 post conference workshop and expo on "Future Directions for Intelligent Information Processing Systems" will be held in Dunedin, New Zealand from 22-23 November 1999. Please see the workshop homepage http://divcom.otago.ac.nz/infosci/kel/iconip99-workshop.htm. An affiliated post-conference workshop is the third Australia-Japan joint Workshop on Intelligent and Evolutionary Systems will be held in Canberra, Australia from 22-23 November 1999. Please see the workshop homepage http://www.cs.adfa.edu.au/conference/aj99/ IMPORTANT DATES Papers due: 16 April 1999 Proposals for tutorials: 1 March 1999 Notification of acceptance: 15 July 1999 Final camera-ready papers due: 31 August 1999 Registration of at least one author of a paper: 31 August 1999 Early registration: 31 August 1999 CONFERENCE CONTACTS, PAPER SUBMISSIONS, CONFERENCE INFORMATION, REGISTRATION FORMS Conference Secretariat ICONIP'99 Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983, Australia phone: +61 8 9470 2552, fax: +61 8 9470 2727, email: iconip99 at at.net.au web: or by post to: Conference Secretariat ICONIP'99 ------------------ Australis Promotions PO Box 1025, Bentley Delivery Centre Perth 6983 Australia ------------------------------------------------------------------------