From herbert.jaeger at ais.fhg.de Mon Jun 3 07:03:40 2002 From: herbert.jaeger at ais.fhg.de (Herbert Jaeger) Date: Mon, 03 Jun 2002 13:03:40 +0200 Subject: Job offer: recurrent neural networks R&D References: Message-ID: <3CFB4D0B.B0652979@ais.fhg.de> RESEARCH POSITION IN APPLICATIONS OF RECURRENT NEURAL NETWORKS ************** apologies for multiple postings ******************* A new research position is open immediately at the Fraunhofer Instiute for Autonomous Intelligent Systems German National Research Center for Information Technology (AIS, www.ais.fraunhofer.de). Field of research: A novel learning method for recurrent neural networks, the "Echo State Network" approach, has been developed at AIS. The method is fast, constructive and outperforms other techniques of nonlinear black-box modeling on benchmarks like the Mackey-Glass chaotic attractor prediction task. The approach is biologically much more plausible than previous methods for recurrent network training. For detailed information, see http://www.ais.fraunhofer.de/INDY/echo_net_research.html. The method will be further developed in AIS towards commercial applications, where fast modeling, filtering and prediction of nonlinear dynamical systems is crucial. Current target areas are in speech recognition, fault monitoring, and nonlinear adaptive control. After the development of demonstrator prototypes, startup enterprises will be founded for the commercial exploitation of the novel methods. AIS solicits applications of researchers/engineers for the development of demonstrator prototypes in one of these areas. The ideal candidate should have a solid background in nonlinear modeling, signal processing and control, the requisite programming skills (Matlab and C/C++), and a keen interest in understanding the dynamics of complex dynamical systems. A business attitude would be an additional assett. When an enterprise is founded, participation of the candidate is desirable. A university degree (master or diploma) is required. Payment will be in agreement with German BAT 2a/1b salary (corresponding to university research assistants/associate researchers). The position will be granted initially for 2 years, with the possibility of further prolongation. AIS is an equal opportunity employer. Please direct inquiries to Dr. Herbert Jaeger Fraunhofer AIS Schloss Birlinghoven D-53754 Sankt Augustin, Germany Phone (+49)-2241-14-2253 Fax (+49)-2241-14-2342 email: herbert.jaeger at gmd.de web: http://www.gmd.de/People/Herbert.Jaeger/ and full applications to Fraunhofer Gesellschaft e.V. Zentrale - Hauptabteilung Personal Schoss Birlinghoven D-53754 Sankt Augustin, Germany In full applications, please refer to the administration number AIS-02-015 of this job offer. From bert at snn.kun.nl Mon Jun 3 11:43:25 2002 From: bert at snn.kun.nl (Bert Kappen) Date: Mon, 3 Jun 2002 17:43:25 +0200 (MEST) Subject: Dynamical synapses and attractor neural networks Message-ID: Dear all, I would like to announce the following paper, that will appear in one of the forthcoming issues of Neural Computation: Associative Memory with dynamic synapses Lovorka Pantic, Joaquin Torres, Bert Kappen and Stan Gielen We have examined a role of dynamic synapses in the stochastic Hopfield-like network behaviour. Our results demonstrate an appearance of a novel phase characterized by quick transitions from one memory state to another. The network is able to retrieve memorized patterns corresponding to classical ferromagnetic states but switches between memorized patterns with an intermittent type of behaviour. This phenomenon might reflect the flexibility of real neural systems and their readiness to receive and respond to novel and changing external stimuli. www.snn.kun.nl/~bert/nc2002.pdf or www.snn.kun.nl/~bert/#biological Bert Kappen SNN University of Nijmegen tel: +31 24 3614241 fax: +31 24 3541435 URL: www.snn.kun.nl/~bert From choe at cs.tamu.edu Sun Jun 16 22:50:46 2002 From: choe at cs.tamu.edu (Yoonsuck - Choe) Date: Sun, 16 Jun 2002 21:50:46 -0500 (CDT) Subject: Dissertation announcement (with demo and software) Message-ID: Dear Colleagues, It is my pleasure to announce the availability of my Ph.D. dissertation, completed last year at the Department of Computer Sciences at the University of Texas at Austin under the supervision of Dr. Risto Miikkulainen. * See below for the URLs of the dissertation, demo, and software. Title: Perceptual Grouping in a Self-Organizing Map of Spiking Neurons Author: Yoonsuck Choe Abstract: Perceptual grouping is the process of identifying the constituents in the visual scene that together form a coherent object. The goal of this thesis is to understand the neural mechanisms of perceptual grouping. The hypotheses are that (1) perceptual grouping is carried out through synchronized firing of neurons representing the same object, and that (2) self-organized lateral connections encoding statistical regularities of the visual environment mediate such a synchronization. A self-organizing neural network of spiking neurons was developed to test these hypotheses in the perceptual grouping task of contour integration. The network self-organized orientation maps and patchy lateral connections similar to those found in the visual cortex, and the contour integration, segmentation, and completion performance measured by the degree of synchrony in neural populations accurately predicted human performance. Such results suggest that synchronized activity can represent perceptual events, and statistical properties of the input can shape the structure of the cortex and the perceptual performance. By providing a computational framework where perceptual performance and neural structure can be compared, the model helps us understand the neural mechanisms of perceptual grouping. Dissertation download: http://www.cs.tamu.edu/faculty/choe/ftp/choe.diss.abstract.html Software: http://www.cs.tamu.edu/faculty/choe/src/pglissom-1.1.tar.gz Demo: http://www.cs.tamu.edu/faculty/choe/demo/pglissom/ Yoonsuck Choe Assistant Professor Deptartment of Computer Science Texas A&M University email: choe at tamu.edu homepage: http://www.cs.tamu.edu/faculty/choe/ From Sebastian_Thrun at heaven.learning.cs.cmu.edu Mon Jun 17 14:51:10 2002 From: Sebastian_Thrun at heaven.learning.cs.cmu.edu (Sebastian Thrun) Date: Mon, 17 Jun 2002 14:51:10 -0400 Subject: NIPS electroninc submission site Message-ID: Dear Connectionists: Just a brief note that NIPS is accepting electronic submissions at http://nips.salk.edu/cgi-bin/nips-2002/NEW_AUTHOR/ca_register_form The submission site will be open until July 1st. Further information on the conference including a link to the page above can be found at NIPS's Web site nips.cc Please direct all inquiries regarding the submission process to me. Sebastian Thrun From Chris.Mellen at gmf.com.au Tue Jun 18 00:59:36 2002 From: Chris.Mellen at gmf.com.au (Chris Mellen) Date: Tue, 18 Jun 2002 14:59:36 +1000 Subject: Research Position, Grinham Managed Funds, Australia Message-ID: <00f501c21684$f4eaadf0$0e80a8c0@researchws01> RESEARCH POSITION GRINHAM MANAGED FUNDS SYDNEY, NSW, AUSTRALIA. Grinham Managed Funds are the Southern Hemisphere's largest hedge fund manager, managing more than US$500 million. We trade over 40 futures markets into 9 countries, 24 hours a day. Currently we are looking to appoint a permanent full-time researcher whose role would be to study financial time-series data using the types of ideas and techniques perhaps more commonly found in the fields of Pattern Recognition, Speech Processing, Machine Learning, Data Compression and the like. It is a requirement for this person to have received a PhD for research already undertaken in one or more of these (or closely related) fields. Prior knowledge or experience of finance is not a prerequisite. The ability to work within a small-team environment is essential. Salary will be commensurate with experience. Bonuses are linked to the individual's and the firm's performance. Individuals who are interested may apply by emailing their Resume to: grinham at gmf.com.au -- Christopher Mellen Grinham Managed Funds, 55 Hume Street, Crows Nest, NSW, 2065, Australia. phone : +61-2-9906 2600 fax : +61-2- 9439 8468 From wolpert at hera.ucl.ac.uk Thu Jun 6 06:23:20 2002 From: wolpert at hera.ucl.ac.uk (Daniel Wolpert) Date: Thu, 6 Jun 2002 11:23:20 +0100 Subject: Two Postdoctoral Fellowships in 1) action observation/imitation & 2) human sensorimotor control Message-ID: <000501c20d44$2ece89a0$58463ec1@hercules> TWO POSTDOCTORAL FELLOWSHIPS In 1) action observation/imitation & 2) human sensorimotor control Institute of Neurology, University College London Two postdoctoral research positions are available in the Sensorimotor Control Laboratory of Professor Daniel Wolpert (wolpert at hera.ucl.ac.uk) in the Sobell Research Department of Motor Neuroscience & Movement Disorders, Institute of Neurology, University College London. The first Postdoctoral Research Fellow will work on a project funded by the McDonnell foundation entitled `Mechanisms of forward thinking and behaviour'. The Fellow will investigate the mechanisms by which the motor system can be used to decode the observed actions of others. The work will involve computational and/or experimental approaches to action observation and imitation. The second Postdoctoral Fellow will work on a project funded by a Wellcome Programme Grant entitled `Computational human sensorimotor control'. The work will involve computational and/or experimental approaches to the planning and learning of skilled actions. Applicants should have a recent doctoral degree in a field relevant to the project. Further details of both posts and laboratory facilities can be found on http://www.wolpertlab.com The positions are available for two years with a starting date from September. Starting salary is up to ?30,726 pa inclusive, depending on experience. Applicants should specify which position they wish to apply for and provide a maximum 1 page statement of research interests relevant to the project, 2 copies of their CV and names of 3 referees to Miss E Bertram, Assistant Secretary (Personnel), Institute of Neurology, Queen Square, London WC1N 3BG (fax: +44 (0)20 7278 5069, email: e.bertram at ion.ucl.ac.uk) by June 28th 2002. Taking Action for Equality From chris at ags.uni-sb.de Thu Jun 6 10:54:14 2002 From: chris at ags.uni-sb.de (Christoph Benzmueller) Date: Thu, 06 Jun 2002 16:54:14 +0200 Subject: STUDENT GRANTS for CALCULEMUS AUTUMN SCHOOL in PISA Message-ID: <3CFF7796.E410E62B@ags.uni-sb.de> ============================================================================ ======== STUDENT GRANTS ============ ======== for CALCULEMUS AUTUMN SCHOOL in PISA ============ ======== http://www.eurice.de/calculemus/autumn-school/ ============ ======== from 23.9.-- 04.10 ============ ============================================================================ Up to 25 STUDENT GRANTS are available for the CALCULEMUS AUTUMN SCHOOL in PISA. They are provided by EU IST Programme (DG Information Society - Future and Emerging Technologies). The student grants cover the registration fees and accommodation at the former monestary Fossabanda in Santa Croce (in a shared room with another student). They amount to approx. 650 EUR. Generally will not fund travel expenses, but in special cases exceptions may be made. To apply for a student grant please send us Curriculum Vitae A short letter of reference A poster submission or extended abstract according to the submission criteria mentioned at the link Student Programmes http://www.eurice.de/calculemus/autumn-school/student_programme.html at the Calculemus Autumn School Webpage The deadline for submission is 1. July. Please send your Applications to (electronic submissions via e-mail are preferred): J?rgen Zimmer AG Deduktionsysteme Fachbereich Informatik (FB 14) Geb?ude 36. 1 Raum 212 Universit?t des Saarlandes D-66041 Saarbr?cken Germany Phone: +49 (0) 681 302 5322 FAX: +49 (0) 681 302 5076 E-mail: jzimmer at mathweb.org Regards, Christoph Benzmueller -- |CHRISTOPH BENZMUELLER www.ags.uni-sb.de/~chris | | FR Informatik, Universitaet des Saarlandes | | Postfach 151150, D-66041 Saarbruecken, Germany | |chris at ags.uni-sb.de Phone:+49-681-302-4574 Fax:+49-681-302-5076 | From D.Willshaw at anc.ed.ac.uk Thu Jun 6 10:13:28 2002 From: D.Willshaw at anc.ed.ac.uk (David Willshaw) Date: Thu, 6 Jun 2002 15:13:28 +0100 Subject: Readership/Lectureship in Neuroinformatics Message-ID: <15615.28168.844078.553088@gargle.gargle.HOWL> Readership or Lecturership in Neuroinformatics The Division of Informatics invites applications from candidates of international standing for an appointment to either a Readership or a Lectureship in Neuroinformatics. Neuroinformatics is broadly defined as including computational modelling of neural and neurocognitive processes, neurally-inspired systems including robotics, and software systems and tools to support neuroscience research. Candidates in all these areas are invited to apply. Edinburgh has a long-standing tradition of excellence in all of the constituent disciplines of neuroinformatics and has recently received in excess of ?15M of new research funding in this area. Candidates should be able to demonstrate a world-class research record and both interest and ability in teaching. The appointee will be expected to participate in the development of the new EPSRC/MRC-funded Doctoral Training Centre in Neuroinformatics and associated research groups; to help develop a new curriculum in neuroinformatics; and to develop his/her own research programme. The appointee to a Readership will also be expected to develop collaborative links and joint activities with academia and industry both nationally and internationally, reflecting the development of neuroinformatics as an international subject. Informal enquiries can be made to David Willshaw Telephone: +44 131 650 4404 Email: david at anc.ed.ac.uk Further information can be found at http://www.jobs.ed.ac.uk/ Salary scale: Lecturer ?20,470 - ?32,215 p.a. Reader ?33,820 - ?38,603 p.a. (Salary levels are subject to annual review.) Application Procedure Please quote Ref: 311456 Letters of application should include a curriculum vitae and the names and addresses of 3 referees. Please include fax numbers and email addresses for referees if possible. Applications should be addressed to Professor David Willshaw, Division of Informatics, and sent to arrive not later than 28 June 2002. By post: 5 Forrest Hill, Edinburgh EH1 2QL By fax: +44 (0)131 650 4406 By email: david at anc.ed.ac.uk On-line: http://www.jobs.ed.ac.uk/ From cogsci at cogprints.soton.ac.uk Fri Jun 7 11:04:06 2002 From: cogsci at cogprints.soton.ac.uk (CogSci WWW Account) Date: Fri, 7 Jun 2002 16:04:06 +0100 (BST) Subject: Cogprints Message-ID: I am forwarding this good advice, originally from Dan Sperber to his Pragmatics list. It applies to all the cognitive sciences. -- Stevan Harnad ---------- Forwarded message ---------- From a.silver at ucl.ac.uk Mon Jun 10 14:16:16 2002 From: a.silver at ucl.ac.uk (Angus Silver) Date: Mon, 10 Jun 2002 19:16:16 +0100 Subject: Postdoc Position in London Message-ID: University College London Computational Neurobiology A postdoctoral position is available in the Department of Physiology. The aim of the project is to examine the role synaptic mechanisms play in signal processing in the granule cell layer of the cerebellar cortex using a physiologically realistic network model. Simulations will be closely linked to electrophysiological experiments carried out in Dr Silver?s lab (e.g Mitchell & Silver Nature 404, 2000) and to other labs that make up the European Commission Cerebellum Group (http://www.cerebellum.org). The candidate will have programming skills and an interest in neurobiology. Experience with the Neuron and/or Genesis simulators would be an advantage. The post is funded for 2.3 years at ?28,625-30,736 p.a. Further information is available from Angus Silver (a.silver at ucl.ac.uk). To apply please send CV before June 15th 2002. Working toward equal opportunity. From tewon at salk.edu Mon Jun 10 20:47:53 2002 From: tewon at salk.edu (Te-Won Lee) Date: Mon, 10 Jun 2002 17:47:53 -0700 Subject: PostDoc position at UCSD Message-ID: <002301c210e1$9f214940$0693ef84@redmond.corp.microsoft.com> POSTDOCTORAL FELLOW POSITION Applications are invited for a joint postdoctoral research associate appointment in the Institute for Neural Computation and the Department of Ophthalmology, both at the University of California, San Diego. This research project involves the development and application of machine learning algorithms and statistical analysis to biomedical data. In particular, we are interested in developing novel machine learning algorithms for the diagnosis and prediction of Glaucoma based on visual fields and retinal imaging as well as structural analysis of the retina. Specific information about the project can be found in publications available from the web link below. This research is in close collaboration with Te-Won Lee INC, UCSD (Principal Investigator) and co-PI: Michael Goldbaum, Pamela Sample, Linda Zangwill, Robert Weinreb (UCSD Medical School), Adam Hoover (CES, Clemson) and Terry Sejnowski (INC, UCSD & Salk Institute). Prospective candidates should hold a Ph.D. in computer science or electrical engineering or related fields with expertise in machine learning algorithms, Bayesian networks and statistical analysis. Experience with biomedical data is ideal but not required. Skills in matlab and C++ are required. More information can be found in http://rhythm.ucsd.edu/~tewon http://inc.ucsd.edu http://eyesite.ucsd.edu/ http://www.ces.clemson.edu/~ahoover/stare/ Please send via regular mail or email: A detailed resume, list of publications, 3 references and a research interest description. Te-Won Lee Te-Won Lee, Ph.D. Institute for Neural Computation - MC0523 University of California, San Diego 9500 Gilman Dr. La Jolla, CA 92039-0523, USA 858-534-9662 office 858-534-2014 fax http://rhythm.ucsd.edu/~tewon From erik at bbf.uia.ac.be Tue Jun 11 07:17:39 2002 From: erik at bbf.uia.ac.be (Erik De Schutter) Date: Tue, 11 Jun 2002 13:17:39 +0200 Subject: Symposium: Neuroinformatics: bioinformatics for the neurosciences? Message-ID: Satellite symposium of the 3rd FORUM OF EUROPEAN NEUROSCIENCE: Neuroinformatics: bioinformatics for the neurosciences? July 13, 2002 - 9.00-15.00, Palais des Congres, Paris Supported by the EC Thematic Network on Computational Neuroscience and Neuroinformatics The speakers at this symposium will present the current status and future trends in the fields of bioinformatics and neuroinformatics and speculate on a possible merger of the two fields. Speakers E. De Schutter (University of Antwerp): T. Hannay (Nature Publishing Group, London) J.P. Changeux (Institut Pasteur, Paris) M. Ashburner (Cambridge University) D. Gardner (Cornell University, New York) T. Shimuzu (Cambridge University) K. Friston (Instute of Neurology, London) Program and free registration at: http://www.neuroinf.org/FENSsat.shtml From fet at socrates.Berkeley.EDU Fri Jun 14 13:08:15 2002 From: fet at socrates.Berkeley.EDU (Frederic Theunissen) Date: Fri, 14 Jun 2002 10:08:15 -0700 Subject: Programming position in Computational Neuroscience at UC Berkeley. Message-ID: The laboratories of Frederic Theunissen, Jack Gallant and Yang Dan at UC Berkeley have a position for a scientific programmer. The programmer will co-lead the design, production and release to the neuroscience community of a computer program with the latest methods for estimating stimulus-response functions of sensory neurons and quantifying information transfer. Excellent benefits, good pay and exciting work with three young investigators. More detailed information about the job can be found at: http://hrweb.berkeley.edu/jvl/search.php?ohrjvlid=13767 and by contacting: Frederic Theunissen fet at socrates.berkeley.edu (510) 643-2247 The applications must be submitted directly to the Human Resources Department at UC Berkeley. Click on the "How to apply" link on the web page. The advertisment on the UC Berkeley web page says: Job Listing Number: 06-767-10 First Posted: 6/13/2002 Job Title:Programmer/Analyst II-UCB Department:Psychology - Psychology Research Salary:$44,500.00 - $80,000.00 per year Application Closing Date:Until Filled, Review Begins: 6/27/2002 Job Starts:7/1/2002, Ends:4/30/2006 Days:Mon,Tue,Wed,Thu,Fri Hours:8:00 AM - 5:00 PM Duties: Design, develop, integrate, and maintain a suite of software applications applying linear and nonlinear systems estimation methods to neurophysiological data sets. Develop new programs as well as integrate functional units of existing programs into a single application with extensive GUI. Perform Web-based software distribution and support and basic system administration tasks for two small research groups. Required Qualifications: A minimum of an undergraduate degree in Electrical Engineering, Computer Science, Physics, Statistics, or related field. Experience with advanced programming techniques in C/C++ and Matlab. Experience in software development with extensive mathematical components and GUI. Experience with Unix (Solaris, Linux) and Microsoft environments. Substantial mathematical knowledge and computer skills. Effective communication skills for interacting with a small group of researchers within the University as well as with researchers outside the University who will be using the software. Desired Qualifications: Familiarity with methods and theory in Systems Neuroscience and/or with signal processing, systems analysis, or neural network theory. Familiarity with statistical theory. Notes: This position is full-time through June 30, 2006 with possible extension depending on the success of the program and availability of future NIH funding. From Mikael.Boden at ide.hh.se Wed Jun 12 10:07:37 2002 From: Mikael.Boden at ide.hh.se (Mikael Bodn) Date: Wed, 12 Jun 2002 16:07:37 +0200 Subject: PhD studentship in machine learning and protein feature prediction Message-ID: <01C2122B.45725650.mibo@ide.hh.se> Applications are invited for a fully funded PhD studentship in the area of machine learning methods for protein feature prediction. The project is supported by the Swedish Research School in Genomics and Bioinformatics. PROJECT DESCRIPTION AND AIM The difficulty of in vivo experimentation often hinders production of large biological data sets. Inadequately sized data sets do not easily lend themselves to statistically sound and proper data analysis. Informed choices of representation and algorithm are required for data analysis tools to produce reliable results. This project aims to provide the groundwork for computational analyses of protein structure and function that use the amino acid sequence information, through a thorough study that pinpoints the strengths and weaknesses of different representations and algorithms. The main goal of the research is to answer two questions: What is, both generally and specifically, a good representation of the amino acid sequence when one uses it to predict protein function? What is a good (efficient, robust, and trustable) algorithm to use for this task? We foresee one outcome of the project to be guidelines for what biological questions are reasonable to ask given the amount of available data (or, conversely, how much data is required to answer specific biological questions) and what are good representations and algorithms for answering these questions. The project will make use of real data as gathered in collaborating projects to re-evaluate theoretical findings, and to contribute to the understanding of the biological phenomena even if only sparse data sets are available. COMPETENCE The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science, statistics, mathematics or other related area with knowledge and interest in machine learning, neural networks and bioinformatics. Programming skills (e.g. MATLAB and Java) are desired. CONTACT INFORMATION Dr. Thorsteinn R?gnvaldsson, denni at ide.hh.se Dr. Mikael Bod?n, mikael.boden at ide.hh.se Halmstad University School of Information Science, Computer- and Electrical Engineering Halmstad, Sweden. APPLICATION Closing date for applications June 28, 2002. Further details are available from http://www.cmb.gu.se/research_school/. --- Dr. Mikael Bod?n, http://www.hh.se/staff/mibo School of Information Science, Computer and Electrical Engineering Halmstad University, PO Box 823, S-30118 Halmstad, Sweden Tel +46 (0)35 167473, Fax +46 (0)35 120348 From becker at meitner.psychology.mcmaster.ca Tue Jun 18 15:58:28 2002 From: becker at meitner.psychology.mcmaster.ca (S. Becker) Date: Tue, 18 Jun 2002 15:58:28 -0400 (EDT) Subject: Postdoc/Research Associate in computational neuroscience Message-ID: POSTDOCTORAL/RESEARCH ASSOCIATE POSITION IN COMPUTATIONAL NEUROSCIENCE AT MCMASTER UNIVERSITY A position is available for a minimum of 3 years at the Postdoctoral or Research Associate level, to develop computational models of cortical learning, memory and reorganization after noise-induced hearing loss. Data from Jos Eggermont's lab indicates that hearing loss in the auditory periphery (sensori-neural hearing loss) results in altered cortical firing rates, tonotopic map reorganization, and abnormal synchronous activation, and may give rise to tinnitus and other perceptual anomalies. Tinnitus, the false perception of sound in the absence of an external sound source, affects about 30% of the population and 70-85% of the hearing-impaired. Remarkably, human brain imaging studies by Roberts, Pantev and colleagues indicate a high degree of plasticity of sensory representations in primary auditory cortex, suggesting good prospects for rehabilitation. Thus, a major goal in this work will be to develop and test novel rehabilitation strategies that tap into this cortical plasticity, for the treatment of tinnitus and other auditory deficits associated with sensori-neural hearing loss. The position offers exciting opportunities to collaborate with team members who are employing a wide range of behavioural neuroscience techniques including computational modelling, psychoacoustics, brain imaging (EEG, MEG), and electrophysiology. Experience with neural network modelling is required, and knowledge of the auditory system, psychoacoustics and signal processing are highly desirable. The position requires a PhD in Cognitive Science, Computer Science, Electrical Engineering or a related discipline. This research is part of a multi-disciplinary project to study mechanisms and treatment of tinnitus, funded by a $1.1M CIHR grant (NET Programme) to L. Roberts and S. Becker (Psychology, McMaster University), I. Bruce (Electrical and Computer Engnineering, McMaster University), J. Eggermont (Faculty of Medicine, University of Calgary), C. Pantev (Rotman Research Institute, Toronto) and L. Ward (Psychology, University of British Columbia). Interested candidates should send a letter of intention and CV via email with subject line "tinnitus postdoc position" to both Dr. S. Becker (becker at mcmaster.ca) and Dr. I. Bruce (ibruce at ieee.org). McMaster University, in Hamilton Ontario, is one of the leading post-secondary institutions in Canada. It was founded in 1887, and has over 13,000 students and 1000 faculty. Known for its strengths in research and innovation, McMaster has numerous award-winning faculty including Bertram Brockhouse, Nobel Laureate in Physics. The pedestrian-only campus is located adjacent to the scenic Royal Botanical Gardens and Coot's Paradise wilderness area at the western end of Lake Ontario. For further information on the research interests of the team see: www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.psychology.mcmaster.ca/hnplab www.ucalgary.ca/UofC/faculties/medicine/PHBI/faculty/eggermont.html www.rotman-baycrest.on.ca/content/people/profiles/pantev.html http://neuron2.psych.ubc.ca/~lward/people/index.html From kramer at ini.phys.ethz.ch Wed Jun 19 05:52:28 2002 From: kramer at ini.phys.ethz.ch (Jorg Kramer) Date: Wed, 19 Jun 2002 11:52:28 +0200 Subject: Positions available in Neuromorphic Engineering Message-ID: <3D10545C.6F19A1@ini.phys.ethz.ch> The Institute of NeuroInformatics invites applications for the following positions in the field of Neuromorphic Engineering: PhD Student Research Assistant (Postdoc or Engineer/Physicist) The preferred starting date is November 1, 2002, or earlier, but later starting dates will also be considered, if necessary. The project concerns a bio-inspired spike-based neuronal processing and communication framework and is a part of an European Commission project in the "Life-Like Perception Systems" initiative of the "Future and Emerging Technologies" programme. The goal of this project is to develop a modular asynchronous communication system and a set of building blocks to be used within this system. Towards the end of the project, the developed components will be tested in a visual perception-action system. The task will involve conception, simulation, layout and testing of analog and asynchronous digital integrated circuits for the building blocks of the system. Dissemination of the results at international conferences and in progress reports will also be part of the assignment. The work will be conducted at the Institute of NeuroInformatics of the University of Zurich and ETH Zurich, which is located on the Irchel campus in Zurich. The institute provides an interdisciplinary environment that includes a group of experienced analog VLSI designers. The research assistant is expected to have some expertise in analog VLSI design, while for the PhD student this is not a prerequisite. The research assistant position is funded for three years and the PhD position is funded for four years. Applicants should have a strong interest in bio-inspired hardware engineering and the ability to work independently, but in coordination with an international team. A good command of English is therefore required. For application or further information, please contact: Shih-Chii Liu, phone: ++41-1-6353047, e-mail: shih at ini.phys.ethz.ch\\ Jorg Kramer, phone: ++41-1-6353039, e-mail: kramer at ini.phys.ethz.ch From terry at salk.edu Wed Jun 19 15:07:11 2002 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 19 Jun 2002 12:07:11 -0700 (PDT) Subject: NEURAL COMPUTATION 14:8 In-Reply-To: <200206010019.g510Jd343973@purkinje.salk.edu> Message-ID: <200206191907.g5JJ7BP78938@remak.salk.edu> Neural Computation - Contents - Volume 14, Number 8 - August 1, 2002 ARTICLE Training products of Experts by Minimizing Contrastive Divergence Geoffrey E. Hinton LETTERS Dynamic Approximation of Spatiotemporal Receptive Fields in Nonlinear Neural Field Models Thomas Wennekers Independent Components of Magnetoencephalography: Localization Akaysha C. Tang, Barak A. Pearlmutter, Natalie A Malaszenko Dan B. Phung and Bethany C. Reeb Robust Blind Source Separation by Beta Divergence Minami Mihoko and Shinto Eguchi Joint Entropy Maximization in Kernel-Based Topographic Maps Marc M. Van Hulle Sufficient Conditions for the Error Backflow Convergence in Dynamical Recurrent Neural Networks Alex Aussem Bayesian A* Tree Search with Expected O(N) Node Expansions: Applications to Road Tracking James M. Coughlan and A. L. Yuille Training v-Support Vector Regression: Theory and Algorithms Chih-Chung Chang and Chih-Jen Lin On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario Katsuyuki Hagiwara ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2002 - VOLUME 14 - 12 ISSUES USA Canada* Other Countries Student/Retired $60 $64.20 $108 Individual $88 $94.16 $136 Institution $506 $451.42 $554 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From Zoubin at gatsby.ucl.ac.uk Wed Jun 19 20:41:21 2002 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 20 Jun 2002 01:41:21 +0100 (BST) Subject: Two Postdocs: Machine Learning / Bioinformatics - London / California Message-ID: <200206200041.BAA18669@cajal.gatsby.ucl.ac.uk> ** Two Postdoctoral Fellowships ** ** Machine Learning and Bioinformatics ** We are seeking two highly creative and motivated postdoctoral research fellows to participate in the following research area: Development of Bayesian network models and machine learning methods for protein fold recognition. This project is part of a collaborative effort to develop a community resource to enable the emerging science of structural genomics. Candidates should have a Ph.D. in computational biology, computer science, machine learning, theoretical physics, applied mathematics, or a similar quantitative field and a strong interest in molecular biology. The positions will be based in London (position 1) and Southern California (position 2), and are described below. The positions form part of a collaborative research project between KGI, UCL, UCSD and the Burnham Institute. Position 1 ---------- Gatsby Computational Neuroscience Unit University College London (UCL) http://www.gatsby.ucl.ac.uk/ Working with Dr Zoubin Ghahramani on advanced inference and learning methods in Bayesian networks and other statistical machine learning methods with applicability to protein fold recognition. The Gatsby Unit at UCL is located in a quiet square in the heart of London, providing an ideal setting for research in computational neuroscience and machine learning in an exciting urban environment. Prospective candidates should apply with a cover letter, CV, and names and email addresses of 2-3 referees. This should be sent by email to: admin at gatsby.ucl.ac.uk, preferably using plain text, postscript or pdf formats only. Closing Date for applications: 31 July 2002. The position is available immediately and the duration of the appointment is 2 years with possibility of renewal. Position 2 ---------- Keck Graduate Institute in Applied Life Sciences (KGI) http://www.kgi.edu/home.html Working with Dr David Wild on the application of machine learning methods to protein fold recognition. KGI is located 35 miles east of Los Angeles, at the foot of the San Gabriel Mountains. Its campus is contiguous with those of the other Claremont Colleges, which together with surrounding educational institutions in Southern California provide a rich intellectual and cultural environment. Prospective candidates should apply with a cover letter and CV, and ask for at least two letters of recommendation to be sent to Dr. David Wild at: Keck Graduate Institute, 535 Watson Drive, Claremont, CA 91711. Email: david_wild at kgi.edu. The position is open immediately and applications will be reviewed as they are received until the position is filled. From harnad at ecs.soton.ac.uk Wed Jun 19 19:36:38 2002 From: harnad at ecs.soton.ac.uk (Stevan Harnad) Date: Thu, 20 Jun 2002 00:36:38 +0100 (BST) Subject: BBS Call for Commentators: Carruthers; The cognitive functions of language (fwd) Message-ID: ---------- Forwarded message ---------- From commentaries at bbsonline.org Wed Jun 19 13:31:05 2002 From: commentaries at bbsonline.org (Behavioral & Brain Sciences) Date: Wed, 19 Jun 2002 13:31:05 -0400 Subject: BBS Call for Commentators: Carruthers; The cognitive functions of language Message-ID: Below is a link to the forthcoming BBS target article The cognitive functions of language by Peter Carruthers http://www.bbsonline.org/Preprints/Carruthers/Referees/ 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. If you are interested in submitting a commentary on this paper, or would like to suggest someone else as a potential commentator on this paper, please read on. 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, pleas= e reply by EMAIL within within three (3) weeks to: calls at bbsonline.org The Calls are sent to over 10,000 researchers in our database, so there is no expectation (indeed, it would be calamitous) that each recipient should comment on every occasion! Hence there is no need to reply except if you wish to comment, or to suggest someone to comment. If you are not a BBS Associate, please approach a current BBS Associate (there are currently over 10,000 worldwide) who is familiar with your work to nominate you. All past BBS authors, referees and commentators are eligible to become BBS Associates. An electronic list of BBS Associates (1978-2000) is available at this location to help you select a name: http://www.bbsonline.org/Instructions/assoclist.html If no current BBS Associate knows your work, please send us your Curriculum Vitae and BBS will circulate it to appropriate Associates to ask whether they would be prepared to nominate you. (In the meantime, your name, address and email address will be entered into our database as an unaffiliated investigator.) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D IMPORTANT To help us put together a balanced list of commentators, please indicate the relevant expertise you would bring to bear on the paper, and what aspect of the article you would anticipate commenting upon. To help you decide whether you would be an appropriate commentator for this article, an electronic draft is retrievable from the online BBSPrints Archive, at the URL proceeding the abstract below. _______________________________________________________________________ The cognitive functions of language Peter Carruthers Department of Philosophy, University of Maryland, College Park, MD 20742 KEYWORDS: cognitive evolution, conceptual module, consciousness, domain-general, inner speech, logical form (LF), language, thought. ABSTRACT: This paper explores a variety of different versions of the thesis that natural language is involved in human thinking. It distinguishes amongst strong and weak forms of this thesis, dismissing some as implausibl= y strong and others as uninterestingly weak. Strong forms dismissed include the view that language is conceptually necessary for thought (endorsed by many philosophers) and the view that language is de facto the medium of all human conceptual thinking (endorsed by many philosophers and social scientists). Weak forms include the view that language is necessary for the acquisition of many human concepts, and the view that language can serve to scaffold human thought processes. The paper also discusses the thesis that language may be the medium of conscious propositional thinking, but argues that this cannot be its most fundamental cognitive role. The idea is then proposed that natural language is the medium for non-domain-specific thinking, serving to integrate the outputs of a variety of domain-specific conceptual faculties (or central-cognitive =91quasi-modules=92). Recent experimental evidence in support of this idea is reviewed, and the implications of the idea are discussed, especially for our conception of th= e architecture of human cognition. Finally, some further kinds of evidence which might serve to corroborate or refute the hypothesis are mentioned. Th= e overall goal of the paper is to review a wide variety of accounts of the cognitive function of natural language, integrating a number of different kinds of evidence and theoretical consideration in order to propose and elaborate the most plausible candidate. http://www.bbsonline.org/Preprints/Carruthers/Referees/ =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D *** SUPPLEMENTARY ANNOUNCEMENT *** (1) Call for Book Nominations for BBS Multiple Book Review In the past, Behavioral and Brain Sciences (BBS) 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!). *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Please note: Your email address has been added to our user database for Calls for Commentators, the reason you received this email. If you do not wish to receive further Calls, please feel free to change your mailshot status through your User Login link on the BBSPrints homepage, using your username and password above: http://www.bbsonline.org/ *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Ralph BBS ------------------------------------------------------------------- Ralph DeMarco Editorial Coordinator Behavioral and Brain Sciences Journals Department Cambridge University Press 40 West 20th Street New York, NY 10011-4211 UNITED STATES bbs at bbsonline.org http://bbsonline.org Tel: +001 212 924 3900 ext.374 Fax: +001 212 645 5960 ------------------------------------------------------------------- From zoubin-nips at gatsby.ucl.ac.uk Wed Jun 19 19:24:33 2002 From: zoubin-nips at gatsby.ucl.ac.uk (zoubin-nips@gatsby.ucl.ac.uk) Date: Thu, 20 Jun 2002 00:24:33 +0100 Subject: Reminder: NIPS Call for Papers / Workshops / Demos Message-ID: <200206192324.AAA18128@cajal.gatsby.ucl.ac.uk> __________________________________________________________________ CALL FOR PAPERS / WORKSHOPS / DEMONSTRATIONS Neural Information Processing Systems Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada http://nips.cc _________________________________________________________________ Summary of Key Dates Paper Submission Deadline: JULY 1, 2002 Workshop Proposal Deadline: August 9, 2002 Demonstration Proposal Deadline: August 9, 2002 Conference: December 9-14, 2002 Submissions are solicited for the sixteenth meeting of an interdisciplinary conference, which brings together cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The conference will include invited talks as well as oral and poster presentations of refereed papers. It is single track and highly selective. Preceding the main conference will be one day of tutorial presentations (Dec.9), and following it there will be two days of focused workshops on topical issues at Whistler/Blackcomb ski resort (Dec.13-14). **** NIPS*2002 INVITED SPEAKERS **** Hugh Durrant-Whyte, University of Sydney Information flow in sensor networks Paul Glimcher, New York University Decisions, Uncertainty and the Brain: Neuroeconomics Deborah Gordon, Stanford University Ants at Work David Heeger, Stanford University Neural Correlates of Perception and Attention Andrew W. Moore, Carnegie Mellon University Statistical Data Mining Pietro Perona, Caltech Learning visual categories **** NIPS*2002 TUTORIAL SPEAKERS **** Martin Cooke Computational auditory scene analysis in listeners and machines Richard M. Karp Mathematical, Statistical and Algorithmic Challenges from Genomics and Molecular Biology Michael Kearns Computational game theory Andrew McCallum Information extraction from the world wide web Sebastian Seung Neural integrators Yair Weiss, Jianbo Shi & Serge Belongie Eigenvector methods for clustering and image segmentation Please visit http://nips.cc for further details. From georg at ai.univie.ac.at Thu Jun 20 13:52:15 2002 From: georg at ai.univie.ac.at (Georg Dorffner) Date: Thu, 20 Jun 2002 19:52:15 +0200 Subject: Positions in process modeling Message-ID: <3D12164F.B19747DD@ai.univie.ac.at> The Neural Computation Group at the Austrian Research Institute for Artificial Intelligence (http://www.ai.univie.ac.at/oefai/nn/) has 1-2 openings for a position in System identification and data modeling in production processes This position will be open starting Sep. 1, 2002 and will be limited to approximately 3 years. Work will be done as part of the EU-commission funded project MONOTONE (Automatic Quality Control in Industrial Printing). This is an application-oriented research project aiming at improving the production process in the ceramics industry. The goal is to model aspects of the production process using semi-parametric methods such as neural networks or Gaussian mixture models, and to solve several pattern recognition problems. Rule-based methods will also be applied, where appropriate. Data will be supplied by several industrial partners. Required background: - know-how in neural computation and/or advanced statistics and data modeling - alternatively, know-how in rule-based and model- based process modeling - academic degree in technical major (e.g. computer science, electrical engineering, mathematics) - research experience - motivation for and interest in industrial research and development - basic programming skills - proficiency in English Desired background - publications related to the field - experience with Matlab - basic understanding of German A Ph.D. degree is not required but will be prefered. Interested applicants should apply by email, mail or fax, no later than July 31, 2002, at the address below. Include a short vita, a list of publications, and any other information demonstrating your qualification. Send applications to: Georg Dorffner Austrian Research Institute for Artificial Intelligence Schottengasse 3 A-1010 Vienna fax: +43-1-4277-9631 email: georg at ai.univie.ac.at [use the keyword MONOTONE in the subject line!] From dominey at isc.cnrs.fr Thu Jun 20 10:55:49 2002 From: dominey at isc.cnrs.fr (Peter Ford Dominey) Date: Thu, 20 Jun 2002 16:55:49 +0200 Subject: POSTDOCTORAL POSITION IN COGNITIVE SCIENCE OF LANGUAGE Message-ID: <3.0.5.32.20020620165549.0096c650@nimbus.isc.cnrs.fr> POSTDOCTORAL POSITION IN COGNITIVE SCIENCE OF LANGUAGE A postdoctoral research opportunity in the Sequential Cognition and Language group (directed by Peter F. Dominey) at the Institut des Sciences Cognitives (Lyon France) is available immediately, to investigate the role of multiple-cue integration in language acquisition across different languages. The project (http://cnl.psych.cornell.edu/mcila) is funded by the Human Frontiers Science Program and involves four closely interacting research teams in France (Peter Dominey, Institut des Sciences Cognitives), the US (Morten Christiansen, Cornell University), the UK (Nick Chater, University of Warwick), and Japan (Mieko Ogura, Tsurumi University). MULTIPLE-CUE INTEGRATION IN LANGUAGE ACQUISITION: MECHANISMS AND NEURAL CORRELATES How do children acquire the subtle and complex structure of their native language with such remarkable speed and reliability, and with little direct instruction? Recent computational and acoustic analyses of language addressed to children indicate that there are rich cues to linguistic structure available in the child's input. Moreover, evidence from developmental psycholinguistics shows that infants are sensitive to many sound-based (phonological) and intonational (prosodic) cues in the input - cues that may facilitate language acquisition. Although this research indicates that linguistic input is rich with possible cues to linguistic structure, there is an important caveat: the cues are only partially reliable and none considered alone provide an infallible bootstrap into language. To acquire language successfully, it seems that the child needs to integrate a great diversity of multiple probabilistic cues to linguistic structure in an effective way. Our research program aims to provide a rigorous cross-linguistic test of the hypothesis that multiple-cue integration is crucial for the acquisition of syntactic structure. The research has four interrelated strands: 1) Computational and acoustic analyses of child-directed speech. 2) Psycholinguistic and artificial language learning experiments. 3) Computational modeling using neural networks and statistical learning methods. 4) Event-related potential (ERP) studies. Specifics for the Lyon Post-Doctoral Position: The selected researcher will participate in this HFSP funded project addressing aspects of language acquisition through simulation, behavioral and brain imagery (ERP) studies. The position will involve: 1. Statistical and acoustic analysis of natural language corpora 2. Participation in neural network simulation of language acquisition processes based on the preceding analysis. An example of a this type of approach can be found in: Dominey PF, Ramus F (2000) Neural network processing of natural language: I. Sensitivity to serial, temporal and abstract structure of language in the infant. Language and Cognitive Processes, 15(1) 87-127 3. Testing of human subjects in artificial language learning experiments. The ideal candidate will have the following qualifications, though candidates with a subset of these will also be considered: 1. A PhD in a related discipline (linguistics/psycholinguistics, computer science, computational neuroscience, cognitive science). 2. Familiarity with the Childes language database and associated analysis tools, and/or experience/interest in computational aspects of language acquisition. 3. Native French, and fluent English. 4. Some computational background, with experience in the Linux/Unix C environment, and in cognitive neuroscience simulation. Interested candidates should send a letter of intention, a CV and three letters of recommendation to Peter F. Dominey at the address below. Applications will continue to be accepted until the position is filled. The position is for one to two years. In addition to salary, funds are available for travel to conferences and meetings between research teams. The position does not carry any special citizen requirements. Peter Ford Dominey, Ph.D. Institut des Sciences Cognitives CNRS UMR 5015 67, Boulevard Pinel 69675 BRON Cedex FRANCE Telephone: 04 37 91 12 12 Direct line: 04 37 91 12 66 FAX: 04 37 91 12 10 email: dominey at isc.cnrs.fr WEB: http://www.isc.cnrs.fr/dom/dommenu.htm From giacomo at ini.phys.ethz.ch Fri Jun 21 06:50:49 2002 From: giacomo at ini.phys.ethz.ch (Giacomo Indiveri) Date: Fri, 21 Jun 2002 12:50:49 +0200 Subject: Open Positions announcement Message-ID: <3D130509.5070203@ini.phys.ethz.ch> Sorry for cross-postings.... Dear Collegues, please forward this announcement to your students and collegues. Best regards, giacomo ====================================================================== Open positions The Institute of NeuroInformatics invites applications for the following positions in the field of Neuromorphic Engineering: PhD Student Research Assistant (Postdoc or Engineer/Physicist) The preferred starting date is October 1, 2002, but later starting dates can be considered. The researchers will be involved in a multi-disciplinary project, funded by the European Community (EU Open FET IST-2001-38099): Attend-to-learn and learn-to-attend with neuromorphic VLSI (ALAVLSI). Note that this project is separate from the CAVIAR project, which is also currently advertising open positions at the Institute of Neuroinformatics. The ALAVLSI project will develop a general architecture for memory-guided attention using both software modelling and neuromorphic VLSI technology. The architecture will be tested on natural visual and auditory stimuli and its performance compared to human observers. These stimuli will utilise a restricted feature space that is small enough to be tractable for VLSI technology yet rich enough for psychophysical tests of attention and learning in human observers/listeners. The task will involve both software modeling of neural networks, and simulation, layout and testing of analog and asynchronous digital integrated circuits for the building blocks of the system. Dissemination of the results at international conferences and in progress reports will also be part of the assignment. The work will be conducted at the Institute of NeuroInformatics of the University of Zurich and ETH Zurich, which is located on the Irchel campus in Zurich. The institute provides an interdisciplinary environment that includes a group of experienced analog VLSI designers. The research assistant is expected to have some expertise in analog VLSI design, while for the PhD student this is not a prerequisite. Both positions are funded for three years. Applicants should have a strong interest in bio-inspired hardware engineering and the ability to work independently, but in coordination with an international team. A good command of English is therefore required. For application or further information, please contact: Giacomo Indiveri, phone: ++41-1-6353024, e-mail: giacomo at ini.phys.ethz.ch ======================================================================= From masulli at disi.unige.it Sat Jun 22 16:04:45 2002 From: masulli at disi.unige.it (Francesco Masulli) Date: Sat, 22 Jun 2002 20:04:45 +0000 Subject: Deadline extention: July 1st, 2002 - School on ENSEMBLE METHODS FOR LEARNING MACHINES Vietri sul Mare 22-28 September 2002 Message-ID: <02062220044501.03523@portofino.disi.unige.it> Dear All, Due to many requests, arrived to the Directors of the Course, to extend the deadline of applications, the deadline has been extended up to July 1st. There would not be other extentions. Best regards Francesco Masulli ================================================================ 7th Course of the "International School on Neural Nets Eduardo R. Caianiello" on ENSEMBLE METHODS FOR LEARNING MACHINES IIASS -Vietri sul Mare (Salerno)-ITALY 22-28 September 2002 web page: http://www.iiass.it/school2002 JOINTLY ORGANIZED BY IIASS-International Institute EMFCSC-Ettore Majorana Foundation for Advanced Scientifc Studies and E.R. Caianiello, Center for Scientific Culture, Vietri sul Mare (SA) Italy Erice (TR) Italy AIMS In the last decade, ensemble methods have shown to be effective in many application domains and constitute one of the main current directions in Machine Learning research. This school will address from a theoretical and empirical view point, several important questions concerning the combination of Learning Machines. In particular, different approaches to the problem which have been proposed in the context of Machine Learning, Neural Networks, and Statistical Pattern Recognition will be discussed. Moreover, a special stress will be given to theoretical and practical tools to develop ensemble methods and evaluate their applications on real-world domains, such as Remote Sensing, Bioinformatics and Medical field. SPONSORS GNCS-Gruppo Nazionale per il Calcolo Scientifico IEEE-Neural Networks Council INNS-International Neural Network Society SIREN-Italian Neural Networks Society University of Salerno,Italy DIRECTORS OF THE COURSE DIRECTORS OF THE SCHOOL Nathan Intrator (Israel and USA) Michael Jordan (USA) Francesco Masulli (Italy) Maria Marinaro (Italy) LECTURERS Leo Breiman, University of California at Berkeley, California, USA Lorenzo Bruzzone, University of Trento, Trento, Italy Thomas G. Dietterich, Oregon State University, Oregon, USA Cesare Furlanello, Istituto per la Ricerca Scientifica e Tecnologica, Trento, Italy Giuseppina C. Gini, Politecnico di Milano, Milano, Italy Tin Kam Ho, Bell Laboratories, New Jersey, USA Nathan Intrator, Tel Aviv Univ., Israel and Brown Univ., Providence, Rhode Island, USA Ludmila I. Kuncheva, University of Wales, Bangor, UK Francesco Masulli, University of Pisa, Italy Stefano Merler, Istituto per la Ricerca Scientifica e Tecnologica, Trento, Italy Fabio Roli, University of Cagliari, Cagliari, Italy Giorgio Valentini, University of Genova, Italy PLACE International Institute for Advanced Scientific Studies E.R. Caianiello (IIASS) Via Pellegrino 19, 84019 Vietri sul Mare, Salerno (Italy) POETIC TOUCH Vietri (from "Veteri", its ancient Roman name) sul Mare ("on sea") is located within walking distance from Salerno and marks the beginning of the Amalfi coast. Short rides take to Positano, Sorrento, Pompei, Herculaneum, Paestum, Vesuvius, or by boat, the islands of Capri, Ischia, and Procida. Velia (the ancient "Elea" of Zeno and Parmenide) is a hundred kilometers farther down along the coast. GENERALITIES Recently, driven by application needs, multiple classifier combinations have evolved into a practical and effective solution for real-world pattern recognition tasks. The idea appears in various disciplines (including Machine Learning, Neural Networks, Pattern Recognition, and Statistics) under several names: hybrid methods, combining decisions, multiple experts, mixture of experts, sensor fusion and many more. In some cases, the combination is motivated by the simple observation that classifier performance is not uniform across the input space and different classifiers excel in different regions. Under a Bayesian framework, integrating over expert distribution leads naturally to expert combination. The generalization capabilities of ensembles of learning machines have been interpreted in the framework of Statistical Learning Theory and in the related theory of Large Margin Classifiers. There are several ways to use more than one classifier in a classification problem. A first "averaging" approach consists of generating multiple hypotheses from a single or multiple learning algorithms, and combining them through majority voting or different linear and non linear combinations. A "feature-oriented" approach is based on different methods to build ensembles of learning machines by subdividing the input space (e.g., random subspace methods, multiple sensors fusion, feature transformation fusion). "Divide-and-conquer" approaches isolate the regions in input space on which each classifier performs well, and direct new input accordingly, or subdivide a complex learning problem in a set of simpler subproblems, recombining them using suitable decoding methods. A "sequential-resampling" approach builds multiple classifier systems using bootstrap methods in order to reduce variance (bagging) or jointly bias and unbiased variance (boosting). There are fundamental questions that need to be addressed for a practical use of this collection of approaches: What are the theoretical tools to interpret possibly in a unified framework this multiplicity of ensemble methods? What is gained and lost in a combination of experts, when is it preferable to alternative approaches? What types of data are best suitable to expert combination? What types of experts are best suited for combinations? What are optimal training methods for experts which are expected to participate in a collective decision? What combination strategies are best suited to a particular problem and to a particular distribution of the data? What are the statistical methods and the appropriate benchmark data to evaluate multiclassifier systems? The school will address some of the above questions from a theoretical and empirical view point and will teach students about this exciting and very promising field using current state of the art data sets for pattern recognition, classification and regression. The main goals of the school are: 1. Offer an overview of the main research issues of ensemble methods from the different and complementary perspectives of Machine Learning, Neural Networks, Statistics and Pattern Recognition. 2. Offer theoretical tools to analyze the diverse approaches, and critically evaluate their applications. 3. Offer practical and theoretical tools to develop new ensemble methods and analyze their application on real-world problems. FORMAT The meeting will follow the usual format of tutorials and panel discussions together with poster sessions for contributed papers. A demo lab with four Linux workstations will be available to the participants for testing and comparing ensemble methods. There will be a network of wireless 11MHz connection available so that students arriving with their laptops and an appropriate wireless communication card can stay connected while at the meeting area. DURATION Participants are expected to arrive in time for the evening meal on Sunday Sept 22th and depart on Saturday Sept 28th. Sessions will take place from Monday Sept 23th to Saturday Sept 28th. PROCEEDINGS The proceedings will be published in the form of a book containing tutorial chapters written by the lecturers and possibly shorter papers from other participants. One free copy of the book will be distributed to each participant after the school. LANGUAGE The official language of the school will be English. POSTER SUBMISSION There will be a poster session for contributed presentations from participants. Proposals consisting of a one page abstract for review by the organizers should be submitted with applications. REGISTRATION FEE Master and PhD Students: 650,00 Euro Academic Participants (govt/univ): 800,00 Euro Industrial Participants: 1.100,00 Euro The fee includes accommodation (3 stars hotel - double room), meals and a copy of the proceedings of the school. Transportation is not included. A supplement of 20 Euro per night should be paid for single room. Members of sponsoring organizations will receive a discount of 50 Euro on the registration fee. A few scholarships are available for students who are otherwise unable to participate at the school. Payment details will be notified with acceptance of applications. ELIGIBILITY The school is open to all suitably qualified scientists. People with few years of experience in the field should include a recommendation letter of their supervisor. APPLICATION PROCEDURE Important Dates: Application deadline: July 1, 2002 Notification of acceptance: July 10, 2002 Registration fee payment deadline: July 20, 2002 School Sept 22-28 2002 Places are limited to a maximum of 60 participants in addition to the lecturers. These will be allocated on a first come, first served basis. ********************************************************************** APPLICATION FORM Title: ............................................................... Family Name: ......................................................... Other Names:.......................................................... Mailing Address (include institution or company name if appropriate): ..................................................................... ..................................................................... ..................................................................... ..................................................................... ..................................................................... Phone:......................Fax:...................................... E-mail: .............................................................. Date of Arrival : .................................................... Date of Departure: ................................................... Are you sending the abstract of a poster? yes/no (delete the alternative which does not apply) Are you applying for a scholarship? yes/no (delete the alternative which does not apply) If yes please include a justification letter for the scholarship request. ***************************************************************** Please send the application form together the recommendation letter by electronic mail to: iiass.vietri at tin.it, subject: summer school; or by fax to: +39 089 761 189 (att.ne Prof. M. Marinaro) or by ordinary mail to the address below: IIASS Via Pellegrino 19, I-84019 Vietri sul Mare (Sa) Italy WEB PAGE OF THE COURSE The web page of the course is http://www.iiass.it/school2002 and will contain all the updates related to the course. At http://www.iiass.it/school2002/ensemble-lab.html a web portal to ENSEMBLE METHODS is in development including pointers to relevant papers, data-bases and software. Contributions to this portal are kindly requested to all researchers involved in this area. Please send all contributions to Giorgio Valentini (valenti at disi.unige.it). FOR FURTHER INFORMATION PLEASE CONTACT: Prof. Francesco Masulli DISI & INFM email: masulli at ge.infm.it Via Dodecaneso 35 fax: +39 010 353 6699 16146 Genova (Italy) tel: +39 010 353 6604 From chris at dcs.bbk.ac.uk Mon Jun 24 07:26:14 2002 From: chris at dcs.bbk.ac.uk (Chris Christodoulou) Date: Mon, 24 Jun 2002 12:26:14 +0100 (BST) Subject: Paper available: A Spiking Neuron Model: Applications and Learning Message-ID: Dear all, I would like to announce the availablility of the following paper that will appear in one of the forthcoming issues of Neural Networks. A preprint of the paper can be downloaded from: http://www.dcs.bbk.ac.uk/~chris/papers/nn_pprnt.pdf or http://www.tech.plym.ac.uk/soc/staff/GuidBugm/pub/nn2002.pdf .............................................................. Title: A Spiking Neuron Model: Applications and Learning Authors: Chris Christodoulou, Guido Bugmann and Trevor G. Clarkson (in press, Neural Networks) Abstract This paper presents a biologically-inspired, hardware-realisable spiking neuron model, which we call the Temporal Noisy-Leaky Integrator (TNLI). The dynamic applications of the model as well as its applications in Computational Neuroscience are demonstrated and a learning algorithm based on postsynaptic delays is proposed. The TNLI incorporates temporal dynamics at the neuron level by modelling both the temporal summation of dendritic postsynaptic currents which have controlled delay and duration and the decay of the somatic potential due to its membrane leak. Moreover, the TNLI models the stochastic neurotransmitter release by real neuron synapses (with probabilistic RAMs at each input) and the firing times including the refractory period and action potential repolarisation. The temporal features of the TNLI make it suitable for use in dynamic time-dependent tasks like its application as a motion and velocity detector system presented in this paper. This is done by modelling the experimental velocity selectivity curve of the motion sensitive H1 neuron of the visual system of the fly. This application of the TNLI indicates its potential applications in artificial vision systems for robots. It is also demonstrated that Hebbian-based learning can be applied in the TNLI for postsynaptic delay training based on coincidence detection, in such a way that an arbitrary temporal pattern can be detected and recognised. The paper also demonstrates that the TNLI can be used to control the firing variability through inhibition; with 80% inhibition to concurrent excitation, firing at high rates is nearly consistent with a Poisson-type firing variability observed in cortical neurons. It is also shown with the TNLI, that the gain of the neuron (slope of its transfer function) can be controlled by the balance between inhibition and excitation, the gain being a decreasing function of the proportion of inhibitory inputs. Finally, in the case of perfect balance between inhibition and excitation, i.e., where the average input current is zero, the neuron can still fire as a result of membrane potential fluctuations. The firing rate is then determined by the average input firing rate. Overall this work illustrates how a hardware-realisable neuron model can capitalise on the unique computational capabilities of biological neurons. Keywords: Spiking Neuron Model; Temporal Noisy-Leaky Integrator; Motion detection; Directional selectivity; postsynaptic delay learning; temporal pattern detection; high firing variability; inhibition. * * * Dr Chris Christodoulou chris at dcs.bbk.ac.uk Chris.Christodoulou at kcl.ac.uk School of Computer Science and Information Systems Birkbeck College, University of London Malet Street, London WC1E 7HX, UK Tel. (+44) 20-7631 6718, Fax (+44) 20-7631 6727 From giro-ci0 at wpmail.paisley.ac.uk Wed Jun 26 09:11:52 2002 From: giro-ci0 at wpmail.paisley.ac.uk (Mark Girolami) Date: Wed, 26 Jun 2002 14:11:52 +0100 Subject: Lecturing Positions Available Message-ID: Dear Connectionists, There are two academic (faculty) positions now available at the University of Paisley and I would be very pleased to hear from subscribers to Connectionists who are interested in these posts. Rgds Mark ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Lecturers in Computing (2)School of Information & Communication TechnologiesFaculty of Communications, Engineering & Science Salary up to 32,537 Two lecturing positions are now available one post will be based at the Paisley Campus of the University and the other will be based at Crichton Campus in Dumfries. You must have at least an upper second class honours degree or equivalent qualification in Computer Science or a related area, a commitment to teaching to a high standard and a good record of research activity. A PhD. Degree is desirable. You will also be expected to participate in administration and to establish research activities in an appropriate area. Informal enquiries to Professor Mark Girolami on 0141 848 3317 E-mail: mark.girolami at paisley.ac.uk. Recruitment packs may be obtained from the Department of Human Resources, University of Paisley, PAISLEY, PA1 2BE (Tel 0141 848 3692 - answering service outwith normal working hours). Closing date: 5 July 2002. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Legal disclaimer -------------------------- The information transmitted is the property of the University of Paisley and is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination and other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer. -------------------------- From mayank at MIT.EDU Wed Jun 26 18:00:53 2002 From: mayank at MIT.EDU (Mayank R Mehta) Date: Wed, 26 Jun 2002 18:00:53 -0400 (EDT) Subject: Paper Available: Role of Experience & Oscillations in Transfomring a Rate Code into a Temporal Code In-Reply-To: <15414.2114.190081.556538@coltrane.ucsf.edu> Message-ID: The following paper can be downloaded from my home page http://www.mit.edu/~mayank This is the final draft of a paper that has appeared in Nature, 417, 741-746 (2002). Summary: While precise spike timing is critical for synaptic plasticity, neither the mechanism for generating a temporal code that establishes such precise spike timing in vivo, nor the effect of experience on such a temporal code have been established. Here we propose a mechanism by which a temporal code can be generated through an interaction between an asymmetric rate code and oscillatory inhibition. Consistent with the predictions of our model the rate and temporal codes of hippocampal pyramidal neurons are highly correlated. Furthermore, the temporal code becomes more robust with experience. The resulting spike timing satisfies the temporal order constraints of Hebbian learning. Thus, oscillations and receptive field asymmetry may play a critical role in temporal sequence learning. ---------------- Apologies for cross-postings. Cheers! -Mayank -----------------------+----------------------------+ Mayank R. Mehta | Email: Mayank at MIT.edu | E18-366, M.I.T. | Phone: 617 252 1841 | 50 Ames St. | FAX: 617 452 4120 | Cambridge, MA 02139 | http://www.mit.edu/~mayank | -----------------------+----------------------------+ From Frederic.Alexandre at loria.fr Fri Jun 28 04:36:01 2002 From: Frederic.Alexandre at loria.fr (Frederic Alexandre) Date: Fri, 28 Jun 2002 10:36:01 +0200 Subject: Two PhD studenship in computer science Message-ID: <3D1C1FF1.DC10560A@loria.fr> The CORTEX team at LORIA/INRIA Lorraine in Nancy, FRANCE is looking for 2 PhD students. http://www.loria.fr/LORIA/LORIA-test/EQUIPES/index.en.php?equipe=Cortex Applications (CV and accompanying letter) are invited for two fully funded Phd studenship, one in the area of neuroscience and modelling, the other in the area of knowledge extraction using neural network. PROJECT 1: In the framework of the MirrorBot European project, between neurosciences and computer science, we aim at designing a biological neural model for the control of an autonomous robot. PROJECT 2: In the framework of an industrial project using extensive geographical data, we aim at designing a neural network for automatic knowledge extraction Details below --- PROJECT 1 In tight connection with CORTEX research themes and taking benefit from our experience in the biologically neural modelling area, we aim at designing a neural networks for the control of an autonomous robot (PeopleBot) in a real but simplified unknown environment. Throughout the many interactions between the robot and the world (objects, positions, colors), we want to study the emergence of semantic during learning. This requires the conception of motor and perceptive systems as well as a sensori-motor coordination system. Moreover, we would like to emphasize the need for different memory systems like declarative memory, procedural memory and working memory. This work will take place in interaction with members of the team and neurobiologists participating to this project. COMPETENCE: The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science or other related area with knowledge and interest in neural networks and modelling. Fluent English (or French) is required. Programming skills (eg C/C++) are desired. CONTACT: Dr. Nicolas Rougier Nicolas.Rougier at loria.fr LORIA laboratory Nancy, France APPLICATION: Closing date for applications: July 15, 2002. --- PROJECT 2 Today, the problem is no longer the need for information but the way to focus on the relevant one and to be able to extract the implicit knowledge contained within. From one hand, internal structures of neural networks are responsible for their good performances in classification and data foraging and in the other hand these internal structures have been hardly studied in terms of information representation and implicit knowledge. Consequently, this work aims at a better understanding of these internal structures and implicit knowledge but also at designing a working model for knowledge extraction. This work will take place in collaboration with the national geological research company (BRGM) in charge of natural ressources in France. COMPETENCE: The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science or other related area with knowledge and interest in neural networks. Fluent English (or French) is required. Programming skills (eg Java, C/C++) and knowledge in AI and statistics are desired. CONTACT: Dr. Laurent Bougrain Laurent.Bougrain at loria.fr LORIA laboratory Nancy, France APPLICATION: Closing date for applications: July 15, 2002. From sue at soc.plym.ac.uk Sun Jun 30 07:34:09 2002 From: sue at soc.plym.ac.uk (Sue Denham) Date: Sun, 30 Jun 2002 12:34:09 +0100 Subject: Research positions in visual and auditory modelling and psychophysics Message-ID: <00b201c2202a$0d77b760$6a4da38d@sue1> Plymouth Institute of Neuroscience Applications are invited for the following positions: 2 postdoctoral research fellowships and 2 research studentships in visual and auditory modelling and psychophysics. The open positions are part of a project, "Attend-to-learn and learn-to-attend with neuromorphic VLSI", funded by the European Community (IST-2001-38099), which combines perceptual, computational, and hardware research. In addition to the University of Plymouth, UK, the project also involves, ETH Zurich and University of Bern, Switzerland, the National Institute of Health, Rome, Italy, Siemens AG, Germany, and the UC Davis, USA. Research at the Plymouth Institute of Neuroscience received a rating of 5 in the 2001 RAE exercise. The aim of the project is to develop a general architecture for memory-guided attention using both software modelling and neuromorphic VLSI technology. The architecture will be tested on natural visual and auditory stimuli and its performance compared to human observers/listeners. Plymouth will be responsible for developing visual and auditory stimuli that are restricted enough to be tractable by an artificial system, yet rich enough for psychophysical tests of attention and learning in human observers/listeners. Specific tasks will include: a.. designing feature spaces that efficiently encode of natural images and sounds b.. conducting psychophysical experiments to characterise attention and learning with novel classes of synthetic images and sounds c.. developing computational models for visual and auditory feature saliency and feature tracking d.. providing performance benchmarks that are suitable for testing the models and hardware components developed by project partners e.. disseminating the results at international conferences. Applicants for the postdoctoral research fellowships should have, or expect to obtain, a Ph.D. in Psychophysics, Computational Modelling, or a related field. Experience in information theoretic analysis is also desirable. Applicants for research studentships should have, or expect to obtain, a good honours degree in Psychology, Neuroscience, Physics, Computing, or related fields. The ideal candidate will possess good analytical, experimental and computational skills. All posts are available from 1 September 2002 for 3 years. Salaries will be internationally competitive and reflect the successful candidate's qualifications. For further information, please contact Prof. Jochen Braun (visual psychophysics and modelling, +44 1752 232 711, achim at pion.ac.uk) or Dr. Sue Denham (auditory psychophysics and modelling, +44 1752 232610, sue at pion.ac.uk). From herbert.jaeger at ais.fhg.de Mon Jun 3 07:03:40 2002 From: herbert.jaeger at ais.fhg.de (Herbert Jaeger) Date: Mon, 03 Jun 2002 13:03:40 +0200 Subject: Job offer: recurrent neural networks R&D References: Message-ID: <3CFB4D0B.B0652979@ais.fhg.de> RESEARCH POSITION IN APPLICATIONS OF RECURRENT NEURAL NETWORKS ************** apologies for multiple postings ******************* A new research position is open immediately at the Fraunhofer Instiute for Autonomous Intelligent Systems German National Research Center for Information Technology (AIS, www.ais.fraunhofer.de). Field of research: A novel learning method for recurrent neural networks, the "Echo State Network" approach, has been developed at AIS. The method is fast, constructive and outperforms other techniques of nonlinear black-box modeling on benchmarks like the Mackey-Glass chaotic attractor prediction task. The approach is biologically much more plausible than previous methods for recurrent network training. For detailed information, see http://www.ais.fraunhofer.de/INDY/echo_net_research.html. The method will be further developed in AIS towards commercial applications, where fast modeling, filtering and prediction of nonlinear dynamical systems is crucial. Current target areas are in speech recognition, fault monitoring, and nonlinear adaptive control. After the development of demonstrator prototypes, startup enterprises will be founded for the commercial exploitation of the novel methods. AIS solicits applications of researchers/engineers for the development of demonstrator prototypes in one of these areas. The ideal candidate should have a solid background in nonlinear modeling, signal processing and control, the requisite programming skills (Matlab and C/C++), and a keen interest in understanding the dynamics of complex dynamical systems. A business attitude would be an additional assett. When an enterprise is founded, participation of the candidate is desirable. A university degree (master or diploma) is required. Payment will be in agreement with German BAT 2a/1b salary (corresponding to university research assistants/associate researchers). The position will be granted initially for 2 years, with the possibility of further prolongation. AIS is an equal opportunity employer. Please direct inquiries to Dr. Herbert Jaeger Fraunhofer AIS Schloss Birlinghoven D-53754 Sankt Augustin, Germany Phone (+49)-2241-14-2253 Fax (+49)-2241-14-2342 email: herbert.jaeger at gmd.de web: http://www.gmd.de/People/Herbert.Jaeger/ and full applications to Fraunhofer Gesellschaft e.V. Zentrale - Hauptabteilung Personal Schoss Birlinghoven D-53754 Sankt Augustin, Germany In full applications, please refer to the administration number AIS-02-015 of this job offer. From bert at snn.kun.nl Mon Jun 3 11:43:25 2002 From: bert at snn.kun.nl (Bert Kappen) Date: Mon, 3 Jun 2002 17:43:25 +0200 (MEST) Subject: Dynamical synapses and attractor neural networks Message-ID: Dear all, I would like to announce the following paper, that will appear in one of the forthcoming issues of Neural Computation: Associative Memory with dynamic synapses Lovorka Pantic, Joaquin Torres, Bert Kappen and Stan Gielen We have examined a role of dynamic synapses in the stochastic Hopfield-like network behaviour. Our results demonstrate an appearance of a novel phase characterized by quick transitions from one memory state to another. The network is able to retrieve memorized patterns corresponding to classical ferromagnetic states but switches between memorized patterns with an intermittent type of behaviour. This phenomenon might reflect the flexibility of real neural systems and their readiness to receive and respond to novel and changing external stimuli. www.snn.kun.nl/~bert/nc2002.pdf or www.snn.kun.nl/~bert/#biological Bert Kappen SNN University of Nijmegen tel: +31 24 3614241 fax: +31 24 3541435 URL: www.snn.kun.nl/~bert From choe at cs.tamu.edu Sun Jun 16 22:50:46 2002 From: choe at cs.tamu.edu (Yoonsuck - Choe) Date: Sun, 16 Jun 2002 21:50:46 -0500 (CDT) Subject: Dissertation announcement (with demo and software) Message-ID: Dear Colleagues, It is my pleasure to announce the availability of my Ph.D. dissertation, completed last year at the Department of Computer Sciences at the University of Texas at Austin under the supervision of Dr. Risto Miikkulainen. * See below for the URLs of the dissertation, demo, and software. Title: Perceptual Grouping in a Self-Organizing Map of Spiking Neurons Author: Yoonsuck Choe Abstract: Perceptual grouping is the process of identifying the constituents in the visual scene that together form a coherent object. The goal of this thesis is to understand the neural mechanisms of perceptual grouping. The hypotheses are that (1) perceptual grouping is carried out through synchronized firing of neurons representing the same object, and that (2) self-organized lateral connections encoding statistical regularities of the visual environment mediate such a synchronization. A self-organizing neural network of spiking neurons was developed to test these hypotheses in the perceptual grouping task of contour integration. The network self-organized orientation maps and patchy lateral connections similar to those found in the visual cortex, and the contour integration, segmentation, and completion performance measured by the degree of synchrony in neural populations accurately predicted human performance. Such results suggest that synchronized activity can represent perceptual events, and statistical properties of the input can shape the structure of the cortex and the perceptual performance. By providing a computational framework where perceptual performance and neural structure can be compared, the model helps us understand the neural mechanisms of perceptual grouping. Dissertation download: http://www.cs.tamu.edu/faculty/choe/ftp/choe.diss.abstract.html Software: http://www.cs.tamu.edu/faculty/choe/src/pglissom-1.1.tar.gz Demo: http://www.cs.tamu.edu/faculty/choe/demo/pglissom/ Yoonsuck Choe Assistant Professor Deptartment of Computer Science Texas A&M University email: choe at tamu.edu homepage: http://www.cs.tamu.edu/faculty/choe/ From Sebastian_Thrun at heaven.learning.cs.cmu.edu Mon Jun 17 14:51:10 2002 From: Sebastian_Thrun at heaven.learning.cs.cmu.edu (Sebastian Thrun) Date: Mon, 17 Jun 2002 14:51:10 -0400 Subject: NIPS electroninc submission site Message-ID: Dear Connectionists: Just a brief note that NIPS is accepting electronic submissions at http://nips.salk.edu/cgi-bin/nips-2002/NEW_AUTHOR/ca_register_form The submission site will be open until July 1st. Further information on the conference including a link to the page above can be found at NIPS's Web site nips.cc Please direct all inquiries regarding the submission process to me. Sebastian Thrun From Chris.Mellen at gmf.com.au Tue Jun 18 00:59:36 2002 From: Chris.Mellen at gmf.com.au (Chris Mellen) Date: Tue, 18 Jun 2002 14:59:36 +1000 Subject: Research Position, Grinham Managed Funds, Australia Message-ID: <00f501c21684$f4eaadf0$0e80a8c0@researchws01> RESEARCH POSITION GRINHAM MANAGED FUNDS SYDNEY, NSW, AUSTRALIA. Grinham Managed Funds are the Southern Hemisphere's largest hedge fund manager, managing more than US$500 million. We trade over 40 futures markets into 9 countries, 24 hours a day. Currently we are looking to appoint a permanent full-time researcher whose role would be to study financial time-series data using the types of ideas and techniques perhaps more commonly found in the fields of Pattern Recognition, Speech Processing, Machine Learning, Data Compression and the like. It is a requirement for this person to have received a PhD for research already undertaken in one or more of these (or closely related) fields. Prior knowledge or experience of finance is not a prerequisite. The ability to work within a small-team environment is essential. Salary will be commensurate with experience. Bonuses are linked to the individual's and the firm's performance. Individuals who are interested may apply by emailing their Resume to: grinham at gmf.com.au -- Christopher Mellen Grinham Managed Funds, 55 Hume Street, Crows Nest, NSW, 2065, Australia. phone : +61-2-9906 2600 fax : +61-2- 9439 8468 From wolpert at hera.ucl.ac.uk Thu Jun 6 06:23:20 2002 From: wolpert at hera.ucl.ac.uk (Daniel Wolpert) Date: Thu, 6 Jun 2002 11:23:20 +0100 Subject: Two Postdoctoral Fellowships in 1) action observation/imitation & 2) human sensorimotor control Message-ID: <000501c20d44$2ece89a0$58463ec1@hercules> TWO POSTDOCTORAL FELLOWSHIPS In 1) action observation/imitation & 2) human sensorimotor control Institute of Neurology, University College London Two postdoctoral research positions are available in the Sensorimotor Control Laboratory of Professor Daniel Wolpert (wolpert at hera.ucl.ac.uk) in the Sobell Research Department of Motor Neuroscience & Movement Disorders, Institute of Neurology, University College London. The first Postdoctoral Research Fellow will work on a project funded by the McDonnell foundation entitled `Mechanisms of forward thinking and behaviour'. The Fellow will investigate the mechanisms by which the motor system can be used to decode the observed actions of others. The work will involve computational and/or experimental approaches to action observation and imitation. The second Postdoctoral Fellow will work on a project funded by a Wellcome Programme Grant entitled `Computational human sensorimotor control'. The work will involve computational and/or experimental approaches to the planning and learning of skilled actions. Applicants should have a recent doctoral degree in a field relevant to the project. Further details of both posts and laboratory facilities can be found on http://www.wolpertlab.com The positions are available for two years with a starting date from September. Starting salary is up to ?30,726 pa inclusive, depending on experience. Applicants should specify which position they wish to apply for and provide a maximum 1 page statement of research interests relevant to the project, 2 copies of their CV and names of 3 referees to Miss E Bertram, Assistant Secretary (Personnel), Institute of Neurology, Queen Square, London WC1N 3BG (fax: +44 (0)20 7278 5069, email: e.bertram at ion.ucl.ac.uk) by June 28th 2002. Taking Action for Equality From chris at ags.uni-sb.de Thu Jun 6 10:54:14 2002 From: chris at ags.uni-sb.de (Christoph Benzmueller) Date: Thu, 06 Jun 2002 16:54:14 +0200 Subject: STUDENT GRANTS for CALCULEMUS AUTUMN SCHOOL in PISA Message-ID: <3CFF7796.E410E62B@ags.uni-sb.de> ============================================================================ ======== STUDENT GRANTS ============ ======== for CALCULEMUS AUTUMN SCHOOL in PISA ============ ======== http://www.eurice.de/calculemus/autumn-school/ ============ ======== from 23.9.-- 04.10 ============ ============================================================================ Up to 25 STUDENT GRANTS are available for the CALCULEMUS AUTUMN SCHOOL in PISA. They are provided by EU IST Programme (DG Information Society - Future and Emerging Technologies). The student grants cover the registration fees and accommodation at the former monestary Fossabanda in Santa Croce (in a shared room with another student). They amount to approx. 650 EUR. Generally will not fund travel expenses, but in special cases exceptions may be made. To apply for a student grant please send us Curriculum Vitae A short letter of reference A poster submission or extended abstract according to the submission criteria mentioned at the link Student Programmes http://www.eurice.de/calculemus/autumn-school/student_programme.html at the Calculemus Autumn School Webpage The deadline for submission is 1. July. Please send your Applications to (electronic submissions via e-mail are preferred): J?rgen Zimmer AG Deduktionsysteme Fachbereich Informatik (FB 14) Geb?ude 36. 1 Raum 212 Universit?t des Saarlandes D-66041 Saarbr?cken Germany Phone: +49 (0) 681 302 5322 FAX: +49 (0) 681 302 5076 E-mail: jzimmer at mathweb.org Regards, Christoph Benzmueller -- |CHRISTOPH BENZMUELLER www.ags.uni-sb.de/~chris | | FR Informatik, Universitaet des Saarlandes | | Postfach 151150, D-66041 Saarbruecken, Germany | |chris at ags.uni-sb.de Phone:+49-681-302-4574 Fax:+49-681-302-5076 | From D.Willshaw at anc.ed.ac.uk Thu Jun 6 10:13:28 2002 From: D.Willshaw at anc.ed.ac.uk (David Willshaw) Date: Thu, 6 Jun 2002 15:13:28 +0100 Subject: Readership/Lectureship in Neuroinformatics Message-ID: <15615.28168.844078.553088@gargle.gargle.HOWL> Readership or Lecturership in Neuroinformatics The Division of Informatics invites applications from candidates of international standing for an appointment to either a Readership or a Lectureship in Neuroinformatics. Neuroinformatics is broadly defined as including computational modelling of neural and neurocognitive processes, neurally-inspired systems including robotics, and software systems and tools to support neuroscience research. Candidates in all these areas are invited to apply. Edinburgh has a long-standing tradition of excellence in all of the constituent disciplines of neuroinformatics and has recently received in excess of ?15M of new research funding in this area. Candidates should be able to demonstrate a world-class research record and both interest and ability in teaching. The appointee will be expected to participate in the development of the new EPSRC/MRC-funded Doctoral Training Centre in Neuroinformatics and associated research groups; to help develop a new curriculum in neuroinformatics; and to develop his/her own research programme. The appointee to a Readership will also be expected to develop collaborative links and joint activities with academia and industry both nationally and internationally, reflecting the development of neuroinformatics as an international subject. Informal enquiries can be made to David Willshaw Telephone: +44 131 650 4404 Email: david at anc.ed.ac.uk Further information can be found at http://www.jobs.ed.ac.uk/ Salary scale: Lecturer ?20,470 - ?32,215 p.a. Reader ?33,820 - ?38,603 p.a. (Salary levels are subject to annual review.) Application Procedure Please quote Ref: 311456 Letters of application should include a curriculum vitae and the names and addresses of 3 referees. Please include fax numbers and email addresses for referees if possible. Applications should be addressed to Professor David Willshaw, Division of Informatics, and sent to arrive not later than 28 June 2002. By post: 5 Forrest Hill, Edinburgh EH1 2QL By fax: +44 (0)131 650 4406 By email: david at anc.ed.ac.uk On-line: http://www.jobs.ed.ac.uk/ From cogsci at cogprints.soton.ac.uk Fri Jun 7 11:04:06 2002 From: cogsci at cogprints.soton.ac.uk (CogSci WWW Account) Date: Fri, 7 Jun 2002 16:04:06 +0100 (BST) Subject: Cogprints Message-ID: I am forwarding this good advice, originally from Dan Sperber to his Pragmatics list. It applies to all the cognitive sciences. -- Stevan Harnad ---------- Forwarded message ---------- From a.silver at ucl.ac.uk Mon Jun 10 14:16:16 2002 From: a.silver at ucl.ac.uk (Angus Silver) Date: Mon, 10 Jun 2002 19:16:16 +0100 Subject: Postdoc Position in London Message-ID: University College London Computational Neurobiology A postdoctoral position is available in the Department of Physiology. The aim of the project is to examine the role synaptic mechanisms play in signal processing in the granule cell layer of the cerebellar cortex using a physiologically realistic network model. Simulations will be closely linked to electrophysiological experiments carried out in Dr Silver?s lab (e.g Mitchell & Silver Nature 404, 2000) and to other labs that make up the European Commission Cerebellum Group (http://www.cerebellum.org). The candidate will have programming skills and an interest in neurobiology. Experience with the Neuron and/or Genesis simulators would be an advantage. The post is funded for 2.3 years at ?28,625-30,736 p.a. Further information is available from Angus Silver (a.silver at ucl.ac.uk). To apply please send CV before June 15th 2002. Working toward equal opportunity. From tewon at salk.edu Mon Jun 10 20:47:53 2002 From: tewon at salk.edu (Te-Won Lee) Date: Mon, 10 Jun 2002 17:47:53 -0700 Subject: PostDoc position at UCSD Message-ID: <002301c210e1$9f214940$0693ef84@redmond.corp.microsoft.com> POSTDOCTORAL FELLOW POSITION Applications are invited for a joint postdoctoral research associate appointment in the Institute for Neural Computation and the Department of Ophthalmology, both at the University of California, San Diego. This research project involves the development and application of machine learning algorithms and statistical analysis to biomedical data. In particular, we are interested in developing novel machine learning algorithms for the diagnosis and prediction of Glaucoma based on visual fields and retinal imaging as well as structural analysis of the retina. Specific information about the project can be found in publications available from the web link below. This research is in close collaboration with Te-Won Lee INC, UCSD (Principal Investigator) and co-PI: Michael Goldbaum, Pamela Sample, Linda Zangwill, Robert Weinreb (UCSD Medical School), Adam Hoover (CES, Clemson) and Terry Sejnowski (INC, UCSD & Salk Institute). Prospective candidates should hold a Ph.D. in computer science or electrical engineering or related fields with expertise in machine learning algorithms, Bayesian networks and statistical analysis. Experience with biomedical data is ideal but not required. Skills in matlab and C++ are required. More information can be found in http://rhythm.ucsd.edu/~tewon http://inc.ucsd.edu http://eyesite.ucsd.edu/ http://www.ces.clemson.edu/~ahoover/stare/ Please send via regular mail or email: A detailed resume, list of publications, 3 references and a research interest description. Te-Won Lee Te-Won Lee, Ph.D. Institute for Neural Computation - MC0523 University of California, San Diego 9500 Gilman Dr. La Jolla, CA 92039-0523, USA 858-534-9662 office 858-534-2014 fax http://rhythm.ucsd.edu/~tewon From erik at bbf.uia.ac.be Tue Jun 11 07:17:39 2002 From: erik at bbf.uia.ac.be (Erik De Schutter) Date: Tue, 11 Jun 2002 13:17:39 +0200 Subject: Symposium: Neuroinformatics: bioinformatics for the neurosciences? Message-ID: Satellite symposium of the 3rd FORUM OF EUROPEAN NEUROSCIENCE: Neuroinformatics: bioinformatics for the neurosciences? July 13, 2002 - 9.00-15.00, Palais des Congres, Paris Supported by the EC Thematic Network on Computational Neuroscience and Neuroinformatics The speakers at this symposium will present the current status and future trends in the fields of bioinformatics and neuroinformatics and speculate on a possible merger of the two fields. Speakers E. De Schutter (University of Antwerp): T. Hannay (Nature Publishing Group, London) J.P. Changeux (Institut Pasteur, Paris) M. Ashburner (Cambridge University) D. Gardner (Cornell University, New York) T. Shimuzu (Cambridge University) K. Friston (Instute of Neurology, London) Program and free registration at: http://www.neuroinf.org/FENSsat.shtml From fet at socrates.Berkeley.EDU Fri Jun 14 13:08:15 2002 From: fet at socrates.Berkeley.EDU (Frederic Theunissen) Date: Fri, 14 Jun 2002 10:08:15 -0700 Subject: Programming position in Computational Neuroscience at UC Berkeley. Message-ID: The laboratories of Frederic Theunissen, Jack Gallant and Yang Dan at UC Berkeley have a position for a scientific programmer. The programmer will co-lead the design, production and release to the neuroscience community of a computer program with the latest methods for estimating stimulus-response functions of sensory neurons and quantifying information transfer. Excellent benefits, good pay and exciting work with three young investigators. More detailed information about the job can be found at: http://hrweb.berkeley.edu/jvl/search.php?ohrjvlid=13767 and by contacting: Frederic Theunissen fet at socrates.berkeley.edu (510) 643-2247 The applications must be submitted directly to the Human Resources Department at UC Berkeley. Click on the "How to apply" link on the web page. The advertisment on the UC Berkeley web page says: Job Listing Number: 06-767-10 First Posted: 6/13/2002 Job Title:Programmer/Analyst II-UCB Department:Psychology - Psychology Research Salary:$44,500.00 - $80,000.00 per year Application Closing Date:Until Filled, Review Begins: 6/27/2002 Job Starts:7/1/2002, Ends:4/30/2006 Days:Mon,Tue,Wed,Thu,Fri Hours:8:00 AM - 5:00 PM Duties: Design, develop, integrate, and maintain a suite of software applications applying linear and nonlinear systems estimation methods to neurophysiological data sets. Develop new programs as well as integrate functional units of existing programs into a single application with extensive GUI. Perform Web-based software distribution and support and basic system administration tasks for two small research groups. Required Qualifications: A minimum of an undergraduate degree in Electrical Engineering, Computer Science, Physics, Statistics, or related field. Experience with advanced programming techniques in C/C++ and Matlab. Experience in software development with extensive mathematical components and GUI. Experience with Unix (Solaris, Linux) and Microsoft environments. Substantial mathematical knowledge and computer skills. Effective communication skills for interacting with a small group of researchers within the University as well as with researchers outside the University who will be using the software. Desired Qualifications: Familiarity with methods and theory in Systems Neuroscience and/or with signal processing, systems analysis, or neural network theory. Familiarity with statistical theory. Notes: This position is full-time through June 30, 2006 with possible extension depending on the success of the program and availability of future NIH funding. From Mikael.Boden at ide.hh.se Wed Jun 12 10:07:37 2002 From: Mikael.Boden at ide.hh.se (Mikael Bodn) Date: Wed, 12 Jun 2002 16:07:37 +0200 Subject: PhD studentship in machine learning and protein feature prediction Message-ID: <01C2122B.45725650.mibo@ide.hh.se> Applications are invited for a fully funded PhD studentship in the area of machine learning methods for protein feature prediction. The project is supported by the Swedish Research School in Genomics and Bioinformatics. PROJECT DESCRIPTION AND AIM The difficulty of in vivo experimentation often hinders production of large biological data sets. Inadequately sized data sets do not easily lend themselves to statistically sound and proper data analysis. Informed choices of representation and algorithm are required for data analysis tools to produce reliable results. This project aims to provide the groundwork for computational analyses of protein structure and function that use the amino acid sequence information, through a thorough study that pinpoints the strengths and weaknesses of different representations and algorithms. The main goal of the research is to answer two questions: What is, both generally and specifically, a good representation of the amino acid sequence when one uses it to predict protein function? What is a good (efficient, robust, and trustable) algorithm to use for this task? We foresee one outcome of the project to be guidelines for what biological questions are reasonable to ask given the amount of available data (or, conversely, how much data is required to answer specific biological questions) and what are good representations and algorithms for answering these questions. The project will make use of real data as gathered in collaborating projects to re-evaluate theoretical findings, and to contribute to the understanding of the biological phenomena even if only sparse data sets are available. COMPETENCE The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science, statistics, mathematics or other related area with knowledge and interest in machine learning, neural networks and bioinformatics. Programming skills (e.g. MATLAB and Java) are desired. CONTACT INFORMATION Dr. Thorsteinn R?gnvaldsson, denni at ide.hh.se Dr. Mikael Bod?n, mikael.boden at ide.hh.se Halmstad University School of Information Science, Computer- and Electrical Engineering Halmstad, Sweden. APPLICATION Closing date for applications June 28, 2002. Further details are available from http://www.cmb.gu.se/research_school/. --- Dr. Mikael Bod?n, http://www.hh.se/staff/mibo School of Information Science, Computer and Electrical Engineering Halmstad University, PO Box 823, S-30118 Halmstad, Sweden Tel +46 (0)35 167473, Fax +46 (0)35 120348 From becker at meitner.psychology.mcmaster.ca Tue Jun 18 15:58:28 2002 From: becker at meitner.psychology.mcmaster.ca (S. Becker) Date: Tue, 18 Jun 2002 15:58:28 -0400 (EDT) Subject: Postdoc/Research Associate in computational neuroscience Message-ID: POSTDOCTORAL/RESEARCH ASSOCIATE POSITION IN COMPUTATIONAL NEUROSCIENCE AT MCMASTER UNIVERSITY A position is available for a minimum of 3 years at the Postdoctoral or Research Associate level, to develop computational models of cortical learning, memory and reorganization after noise-induced hearing loss. Data from Jos Eggermont's lab indicates that hearing loss in the auditory periphery (sensori-neural hearing loss) results in altered cortical firing rates, tonotopic map reorganization, and abnormal synchronous activation, and may give rise to tinnitus and other perceptual anomalies. Tinnitus, the false perception of sound in the absence of an external sound source, affects about 30% of the population and 70-85% of the hearing-impaired. Remarkably, human brain imaging studies by Roberts, Pantev and colleagues indicate a high degree of plasticity of sensory representations in primary auditory cortex, suggesting good prospects for rehabilitation. Thus, a major goal in this work will be to develop and test novel rehabilitation strategies that tap into this cortical plasticity, for the treatment of tinnitus and other auditory deficits associated with sensori-neural hearing loss. The position offers exciting opportunities to collaborate with team members who are employing a wide range of behavioural neuroscience techniques including computational modelling, psychoacoustics, brain imaging (EEG, MEG), and electrophysiology. Experience with neural network modelling is required, and knowledge of the auditory system, psychoacoustics and signal processing are highly desirable. The position requires a PhD in Cognitive Science, Computer Science, Electrical Engineering or a related discipline. This research is part of a multi-disciplinary project to study mechanisms and treatment of tinnitus, funded by a $1.1M CIHR grant (NET Programme) to L. Roberts and S. Becker (Psychology, McMaster University), I. Bruce (Electrical and Computer Engnineering, McMaster University), J. Eggermont (Faculty of Medicine, University of Calgary), C. Pantev (Rotman Research Institute, Toronto) and L. Ward (Psychology, University of British Columbia). Interested candidates should send a letter of intention and CV via email with subject line "tinnitus postdoc position" to both Dr. S. Becker (becker at mcmaster.ca) and Dr. I. Bruce (ibruce at ieee.org). McMaster University, in Hamilton Ontario, is one of the leading post-secondary institutions in Canada. It was founded in 1887, and has over 13,000 students and 1000 faculty. Known for its strengths in research and innovation, McMaster has numerous award-winning faculty including Bertram Brockhouse, Nobel Laureate in Physics. The pedestrian-only campus is located adjacent to the scenic Royal Botanical Gardens and Coot's Paradise wilderness area at the western end of Lake Ontario. For further information on the research interests of the team see: www.science.mcmaster.ca/Psychology/sb.html www.ece.mcmaster.ca/~ibruce/ www.psychology.mcmaster.ca/hnplab www.ucalgary.ca/UofC/faculties/medicine/PHBI/faculty/eggermont.html www.rotman-baycrest.on.ca/content/people/profiles/pantev.html http://neuron2.psych.ubc.ca/~lward/people/index.html From kramer at ini.phys.ethz.ch Wed Jun 19 05:52:28 2002 From: kramer at ini.phys.ethz.ch (Jorg Kramer) Date: Wed, 19 Jun 2002 11:52:28 +0200 Subject: Positions available in Neuromorphic Engineering Message-ID: <3D10545C.6F19A1@ini.phys.ethz.ch> The Institute of NeuroInformatics invites applications for the following positions in the field of Neuromorphic Engineering: PhD Student Research Assistant (Postdoc or Engineer/Physicist) The preferred starting date is November 1, 2002, or earlier, but later starting dates will also be considered, if necessary. The project concerns a bio-inspired spike-based neuronal processing and communication framework and is a part of an European Commission project in the "Life-Like Perception Systems" initiative of the "Future and Emerging Technologies" programme. The goal of this project is to develop a modular asynchronous communication system and a set of building blocks to be used within this system. Towards the end of the project, the developed components will be tested in a visual perception-action system. The task will involve conception, simulation, layout and testing of analog and asynchronous digital integrated circuits for the building blocks of the system. Dissemination of the results at international conferences and in progress reports will also be part of the assignment. The work will be conducted at the Institute of NeuroInformatics of the University of Zurich and ETH Zurich, which is located on the Irchel campus in Zurich. The institute provides an interdisciplinary environment that includes a group of experienced analog VLSI designers. The research assistant is expected to have some expertise in analog VLSI design, while for the PhD student this is not a prerequisite. The research assistant position is funded for three years and the PhD position is funded for four years. Applicants should have a strong interest in bio-inspired hardware engineering and the ability to work independently, but in coordination with an international team. A good command of English is therefore required. For application or further information, please contact: Shih-Chii Liu, phone: ++41-1-6353047, e-mail: shih at ini.phys.ethz.ch\\ Jorg Kramer, phone: ++41-1-6353039, e-mail: kramer at ini.phys.ethz.ch From terry at salk.edu Wed Jun 19 15:07:11 2002 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 19 Jun 2002 12:07:11 -0700 (PDT) Subject: NEURAL COMPUTATION 14:8 In-Reply-To: <200206010019.g510Jd343973@purkinje.salk.edu> Message-ID: <200206191907.g5JJ7BP78938@remak.salk.edu> Neural Computation - Contents - Volume 14, Number 8 - August 1, 2002 ARTICLE Training products of Experts by Minimizing Contrastive Divergence Geoffrey E. Hinton LETTERS Dynamic Approximation of Spatiotemporal Receptive Fields in Nonlinear Neural Field Models Thomas Wennekers Independent Components of Magnetoencephalography: Localization Akaysha C. Tang, Barak A. Pearlmutter, Natalie A Malaszenko Dan B. Phung and Bethany C. Reeb Robust Blind Source Separation by Beta Divergence Minami Mihoko and Shinto Eguchi Joint Entropy Maximization in Kernel-Based Topographic Maps Marc M. Van Hulle Sufficient Conditions for the Error Backflow Convergence in Dynamical Recurrent Neural Networks Alex Aussem Bayesian A* Tree Search with Expected O(N) Node Expansions: Applications to Road Tracking James M. Coughlan and A. L. Yuille Training v-Support Vector Regression: Theory and Algorithms Chih-Chung Chang and Chih-Jen Lin On the Problem in Model Selection of Neural Network Regression in Overrealizable Scenario Katsuyuki Hagiwara ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2002 - VOLUME 14 - 12 ISSUES USA Canada* Other Countries Student/Retired $60 $64.20 $108 Individual $88 $94.16 $136 Institution $506 $451.42 $554 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From Zoubin at gatsby.ucl.ac.uk Wed Jun 19 20:41:21 2002 From: Zoubin at gatsby.ucl.ac.uk (Zoubin Ghahramani) Date: Thu, 20 Jun 2002 01:41:21 +0100 (BST) Subject: Two Postdocs: Machine Learning / Bioinformatics - London / California Message-ID: <200206200041.BAA18669@cajal.gatsby.ucl.ac.uk> ** Two Postdoctoral Fellowships ** ** Machine Learning and Bioinformatics ** We are seeking two highly creative and motivated postdoctoral research fellows to participate in the following research area: Development of Bayesian network models and machine learning methods for protein fold recognition. This project is part of a collaborative effort to develop a community resource to enable the emerging science of structural genomics. Candidates should have a Ph.D. in computational biology, computer science, machine learning, theoretical physics, applied mathematics, or a similar quantitative field and a strong interest in molecular biology. The positions will be based in London (position 1) and Southern California (position 2), and are described below. The positions form part of a collaborative research project between KGI, UCL, UCSD and the Burnham Institute. Position 1 ---------- Gatsby Computational Neuroscience Unit University College London (UCL) http://www.gatsby.ucl.ac.uk/ Working with Dr Zoubin Ghahramani on advanced inference and learning methods in Bayesian networks and other statistical machine learning methods with applicability to protein fold recognition. The Gatsby Unit at UCL is located in a quiet square in the heart of London, providing an ideal setting for research in computational neuroscience and machine learning in an exciting urban environment. Prospective candidates should apply with a cover letter, CV, and names and email addresses of 2-3 referees. This should be sent by email to: admin at gatsby.ucl.ac.uk, preferably using plain text, postscript or pdf formats only. Closing Date for applications: 31 July 2002. The position is available immediately and the duration of the appointment is 2 years with possibility of renewal. Position 2 ---------- Keck Graduate Institute in Applied Life Sciences (KGI) http://www.kgi.edu/home.html Working with Dr David Wild on the application of machine learning methods to protein fold recognition. KGI is located 35 miles east of Los Angeles, at the foot of the San Gabriel Mountains. Its campus is contiguous with those of the other Claremont Colleges, which together with surrounding educational institutions in Southern California provide a rich intellectual and cultural environment. Prospective candidates should apply with a cover letter and CV, and ask for at least two letters of recommendation to be sent to Dr. David Wild at: Keck Graduate Institute, 535 Watson Drive, Claremont, CA 91711. Email: david_wild at kgi.edu. The position is open immediately and applications will be reviewed as they are received until the position is filled. From harnad at ecs.soton.ac.uk Wed Jun 19 19:36:38 2002 From: harnad at ecs.soton.ac.uk (Stevan Harnad) Date: Thu, 20 Jun 2002 00:36:38 +0100 (BST) Subject: BBS Call for Commentators: Carruthers; The cognitive functions of language (fwd) Message-ID: ---------- Forwarded message ---------- From commentaries at bbsonline.org Wed Jun 19 13:31:05 2002 From: commentaries at bbsonline.org (Behavioral & Brain Sciences) Date: Wed, 19 Jun 2002 13:31:05 -0400 Subject: BBS Call for Commentators: Carruthers; The cognitive functions of language Message-ID: Below is a link to the forthcoming BBS target article The cognitive functions of language by Peter Carruthers http://www.bbsonline.org/Preprints/Carruthers/Referees/ 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. If you are interested in submitting a commentary on this paper, or would like to suggest someone else as a potential commentator on this paper, please read on. 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, pleas= e reply by EMAIL within within three (3) weeks to: calls at bbsonline.org The Calls are sent to over 10,000 researchers in our database, so there is no expectation (indeed, it would be calamitous) that each recipient should comment on every occasion! Hence there is no need to reply except if you wish to comment, or to suggest someone to comment. If you are not a BBS Associate, please approach a current BBS Associate (there are currently over 10,000 worldwide) who is familiar with your work to nominate you. All past BBS authors, referees and commentators are eligible to become BBS Associates. An electronic list of BBS Associates (1978-2000) is available at this location to help you select a name: http://www.bbsonline.org/Instructions/assoclist.html If no current BBS Associate knows your work, please send us your Curriculum Vitae and BBS will circulate it to appropriate Associates to ask whether they would be prepared to nominate you. (In the meantime, your name, address and email address will be entered into our database as an unaffiliated investigator.) =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D IMPORTANT To help us put together a balanced list of commentators, please indicate the relevant expertise you would bring to bear on the paper, and what aspect of the article you would anticipate commenting upon. To help you decide whether you would be an appropriate commentator for this article, an electronic draft is retrievable from the online BBSPrints Archive, at the URL proceeding the abstract below. _______________________________________________________________________ The cognitive functions of language Peter Carruthers Department of Philosophy, University of Maryland, College Park, MD 20742 KEYWORDS: cognitive evolution, conceptual module, consciousness, domain-general, inner speech, logical form (LF), language, thought. ABSTRACT: This paper explores a variety of different versions of the thesis that natural language is involved in human thinking. It distinguishes amongst strong and weak forms of this thesis, dismissing some as implausibl= y strong and others as uninterestingly weak. Strong forms dismissed include the view that language is conceptually necessary for thought (endorsed by many philosophers) and the view that language is de facto the medium of all human conceptual thinking (endorsed by many philosophers and social scientists). Weak forms include the view that language is necessary for the acquisition of many human concepts, and the view that language can serve to scaffold human thought processes. The paper also discusses the thesis that language may be the medium of conscious propositional thinking, but argues that this cannot be its most fundamental cognitive role. The idea is then proposed that natural language is the medium for non-domain-specific thinking, serving to integrate the outputs of a variety of domain-specific conceptual faculties (or central-cognitive =91quasi-modules=92). Recent experimental evidence in support of this idea is reviewed, and the implications of the idea are discussed, especially for our conception of th= e architecture of human cognition. Finally, some further kinds of evidence which might serve to corroborate or refute the hypothesis are mentioned. Th= e overall goal of the paper is to review a wide variety of accounts of the cognitive function of natural language, integrating a number of different kinds of evidence and theoretical consideration in order to propose and elaborate the most plausible candidate. http://www.bbsonline.org/Preprints/Carruthers/Referees/ =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D *** SUPPLEMENTARY ANNOUNCEMENT *** (1) Call for Book Nominations for BBS Multiple Book Review In the past, Behavioral and Brain Sciences (BBS) 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!). *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Please note: Your email address has been added to our user database for Calls for Commentators, the reason you received this email. If you do not wish to receive further Calls, please feel free to change your mailshot status through your User Login link on the BBSPrints homepage, using your username and password above: http://www.bbsonline.org/ *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* Ralph BBS ------------------------------------------------------------------- Ralph DeMarco Editorial Coordinator Behavioral and Brain Sciences Journals Department Cambridge University Press 40 West 20th Street New York, NY 10011-4211 UNITED STATES bbs at bbsonline.org http://bbsonline.org Tel: +001 212 924 3900 ext.374 Fax: +001 212 645 5960 ------------------------------------------------------------------- From zoubin-nips at gatsby.ucl.ac.uk Wed Jun 19 19:24:33 2002 From: zoubin-nips at gatsby.ucl.ac.uk (zoubin-nips@gatsby.ucl.ac.uk) Date: Thu, 20 Jun 2002 00:24:33 +0100 Subject: Reminder: NIPS Call for Papers / Workshops / Demos Message-ID: <200206192324.AAA18128@cajal.gatsby.ucl.ac.uk> __________________________________________________________________ CALL FOR PAPERS / WORKSHOPS / DEMONSTRATIONS Neural Information Processing Systems Monday, December 9 -- Saturday December 14, 2002 Vancouver, British Columbia, Canada http://nips.cc _________________________________________________________________ Summary of Key Dates Paper Submission Deadline: JULY 1, 2002 Workshop Proposal Deadline: August 9, 2002 Demonstration Proposal Deadline: August 9, 2002 Conference: December 9-14, 2002 Submissions are solicited for the sixteenth meeting of an interdisciplinary conference, which brings together cognitive scientists, computer scientists, engineers, neuroscientists, physicists, statisticians, and mathematicians interested in all aspects of neural and statistical processing and computation. The conference will include invited talks as well as oral and poster presentations of refereed papers. It is single track and highly selective. Preceding the main conference will be one day of tutorial presentations (Dec.9), and following it there will be two days of focused workshops on topical issues at Whistler/Blackcomb ski resort (Dec.13-14). **** NIPS*2002 INVITED SPEAKERS **** Hugh Durrant-Whyte, University of Sydney Information flow in sensor networks Paul Glimcher, New York University Decisions, Uncertainty and the Brain: Neuroeconomics Deborah Gordon, Stanford University Ants at Work David Heeger, Stanford University Neural Correlates of Perception and Attention Andrew W. Moore, Carnegie Mellon University Statistical Data Mining Pietro Perona, Caltech Learning visual categories **** NIPS*2002 TUTORIAL SPEAKERS **** Martin Cooke Computational auditory scene analysis in listeners and machines Richard M. Karp Mathematical, Statistical and Algorithmic Challenges from Genomics and Molecular Biology Michael Kearns Computational game theory Andrew McCallum Information extraction from the world wide web Sebastian Seung Neural integrators Yair Weiss, Jianbo Shi & Serge Belongie Eigenvector methods for clustering and image segmentation Please visit http://nips.cc for further details. From georg at ai.univie.ac.at Thu Jun 20 13:52:15 2002 From: georg at ai.univie.ac.at (Georg Dorffner) Date: Thu, 20 Jun 2002 19:52:15 +0200 Subject: Positions in process modeling Message-ID: <3D12164F.B19747DD@ai.univie.ac.at> The Neural Computation Group at the Austrian Research Institute for Artificial Intelligence (http://www.ai.univie.ac.at/oefai/nn/) has 1-2 openings for a position in System identification and data modeling in production processes This position will be open starting Sep. 1, 2002 and will be limited to approximately 3 years. Work will be done as part of the EU-commission funded project MONOTONE (Automatic Quality Control in Industrial Printing). This is an application-oriented research project aiming at improving the production process in the ceramics industry. The goal is to model aspects of the production process using semi-parametric methods such as neural networks or Gaussian mixture models, and to solve several pattern recognition problems. Rule-based methods will also be applied, where appropriate. Data will be supplied by several industrial partners. Required background: - know-how in neural computation and/or advanced statistics and data modeling - alternatively, know-how in rule-based and model- based process modeling - academic degree in technical major (e.g. computer science, electrical engineering, mathematics) - research experience - motivation for and interest in industrial research and development - basic programming skills - proficiency in English Desired background - publications related to the field - experience with Matlab - basic understanding of German A Ph.D. degree is not required but will be prefered. Interested applicants should apply by email, mail or fax, no later than July 31, 2002, at the address below. Include a short vita, a list of publications, and any other information demonstrating your qualification. Send applications to: Georg Dorffner Austrian Research Institute for Artificial Intelligence Schottengasse 3 A-1010 Vienna fax: +43-1-4277-9631 email: georg at ai.univie.ac.at [use the keyword MONOTONE in the subject line!] From dominey at isc.cnrs.fr Thu Jun 20 10:55:49 2002 From: dominey at isc.cnrs.fr (Peter Ford Dominey) Date: Thu, 20 Jun 2002 16:55:49 +0200 Subject: POSTDOCTORAL POSITION IN COGNITIVE SCIENCE OF LANGUAGE Message-ID: <3.0.5.32.20020620165549.0096c650@nimbus.isc.cnrs.fr> POSTDOCTORAL POSITION IN COGNITIVE SCIENCE OF LANGUAGE A postdoctoral research opportunity in the Sequential Cognition and Language group (directed by Peter F. Dominey) at the Institut des Sciences Cognitives (Lyon France) is available immediately, to investigate the role of multiple-cue integration in language acquisition across different languages. The project (http://cnl.psych.cornell.edu/mcila) is funded by the Human Frontiers Science Program and involves four closely interacting research teams in France (Peter Dominey, Institut des Sciences Cognitives), the US (Morten Christiansen, Cornell University), the UK (Nick Chater, University of Warwick), and Japan (Mieko Ogura, Tsurumi University). MULTIPLE-CUE INTEGRATION IN LANGUAGE ACQUISITION: MECHANISMS AND NEURAL CORRELATES How do children acquire the subtle and complex structure of their native language with such remarkable speed and reliability, and with little direct instruction? Recent computational and acoustic analyses of language addressed to children indicate that there are rich cues to linguistic structure available in the child's input. Moreover, evidence from developmental psycholinguistics shows that infants are sensitive to many sound-based (phonological) and intonational (prosodic) cues in the input - cues that may facilitate language acquisition. Although this research indicates that linguistic input is rich with possible cues to linguistic structure, there is an important caveat: the cues are only partially reliable and none considered alone provide an infallible bootstrap into language. To acquire language successfully, it seems that the child needs to integrate a great diversity of multiple probabilistic cues to linguistic structure in an effective way. Our research program aims to provide a rigorous cross-linguistic test of the hypothesis that multiple-cue integration is crucial for the acquisition of syntactic structure. The research has four interrelated strands: 1) Computational and acoustic analyses of child-directed speech. 2) Psycholinguistic and artificial language learning experiments. 3) Computational modeling using neural networks and statistical learning methods. 4) Event-related potential (ERP) studies. Specifics for the Lyon Post-Doctoral Position: The selected researcher will participate in this HFSP funded project addressing aspects of language acquisition through simulation, behavioral and brain imagery (ERP) studies. The position will involve: 1. Statistical and acoustic analysis of natural language corpora 2. Participation in neural network simulation of language acquisition processes based on the preceding analysis. An example of a this type of approach can be found in: Dominey PF, Ramus F (2000) Neural network processing of natural language: I. Sensitivity to serial, temporal and abstract structure of language in the infant. Language and Cognitive Processes, 15(1) 87-127 3. Testing of human subjects in artificial language learning experiments. The ideal candidate will have the following qualifications, though candidates with a subset of these will also be considered: 1. A PhD in a related discipline (linguistics/psycholinguistics, computer science, computational neuroscience, cognitive science). 2. Familiarity with the Childes language database and associated analysis tools, and/or experience/interest in computational aspects of language acquisition. 3. Native French, and fluent English. 4. Some computational background, with experience in the Linux/Unix C environment, and in cognitive neuroscience simulation. Interested candidates should send a letter of intention, a CV and three letters of recommendation to Peter F. Dominey at the address below. Applications will continue to be accepted until the position is filled. The position is for one to two years. In addition to salary, funds are available for travel to conferences and meetings between research teams. The position does not carry any special citizen requirements. Peter Ford Dominey, Ph.D. Institut des Sciences Cognitives CNRS UMR 5015 67, Boulevard Pinel 69675 BRON Cedex FRANCE Telephone: 04 37 91 12 12 Direct line: 04 37 91 12 66 FAX: 04 37 91 12 10 email: dominey at isc.cnrs.fr WEB: http://www.isc.cnrs.fr/dom/dommenu.htm From giacomo at ini.phys.ethz.ch Fri Jun 21 06:50:49 2002 From: giacomo at ini.phys.ethz.ch (Giacomo Indiveri) Date: Fri, 21 Jun 2002 12:50:49 +0200 Subject: Open Positions announcement Message-ID: <3D130509.5070203@ini.phys.ethz.ch> Sorry for cross-postings.... Dear Collegues, please forward this announcement to your students and collegues. Best regards, giacomo ====================================================================== Open positions The Institute of NeuroInformatics invites applications for the following positions in the field of Neuromorphic Engineering: PhD Student Research Assistant (Postdoc or Engineer/Physicist) The preferred starting date is October 1, 2002, but later starting dates can be considered. The researchers will be involved in a multi-disciplinary project, funded by the European Community (EU Open FET IST-2001-38099): Attend-to-learn and learn-to-attend with neuromorphic VLSI (ALAVLSI). Note that this project is separate from the CAVIAR project, which is also currently advertising open positions at the Institute of Neuroinformatics. The ALAVLSI project will develop a general architecture for memory-guided attention using both software modelling and neuromorphic VLSI technology. The architecture will be tested on natural visual and auditory stimuli and its performance compared to human observers. These stimuli will utilise a restricted feature space that is small enough to be tractable for VLSI technology yet rich enough for psychophysical tests of attention and learning in human observers/listeners. The task will involve both software modeling of neural networks, and simulation, layout and testing of analog and asynchronous digital integrated circuits for the building blocks of the system. Dissemination of the results at international conferences and in progress reports will also be part of the assignment. The work will be conducted at the Institute of NeuroInformatics of the University of Zurich and ETH Zurich, which is located on the Irchel campus in Zurich. The institute provides an interdisciplinary environment that includes a group of experienced analog VLSI designers. The research assistant is expected to have some expertise in analog VLSI design, while for the PhD student this is not a prerequisite. Both positions are funded for three years. Applicants should have a strong interest in bio-inspired hardware engineering and the ability to work independently, but in coordination with an international team. A good command of English is therefore required. For application or further information, please contact: Giacomo Indiveri, phone: ++41-1-6353024, e-mail: giacomo at ini.phys.ethz.ch ======================================================================= From masulli at disi.unige.it Sat Jun 22 16:04:45 2002 From: masulli at disi.unige.it (Francesco Masulli) Date: Sat, 22 Jun 2002 20:04:45 +0000 Subject: Deadline extention: July 1st, 2002 - School on ENSEMBLE METHODS FOR LEARNING MACHINES Vietri sul Mare 22-28 September 2002 Message-ID: <02062220044501.03523@portofino.disi.unige.it> Dear All, Due to many requests, arrived to the Directors of the Course, to extend the deadline of applications, the deadline has been extended up to July 1st. There would not be other extentions. Best regards Francesco Masulli ================================================================ 7th Course of the "International School on Neural Nets Eduardo R. Caianiello" on ENSEMBLE METHODS FOR LEARNING MACHINES IIASS -Vietri sul Mare (Salerno)-ITALY 22-28 September 2002 web page: http://www.iiass.it/school2002 JOINTLY ORGANIZED BY IIASS-International Institute EMFCSC-Ettore Majorana Foundation for Advanced Scientifc Studies and E.R. Caianiello, Center for Scientific Culture, Vietri sul Mare (SA) Italy Erice (TR) Italy AIMS In the last decade, ensemble methods have shown to be effective in many application domains and constitute one of the main current directions in Machine Learning research. This school will address from a theoretical and empirical view point, several important questions concerning the combination of Learning Machines. In particular, different approaches to the problem which have been proposed in the context of Machine Learning, Neural Networks, and Statistical Pattern Recognition will be discussed. Moreover, a special stress will be given to theoretical and practical tools to develop ensemble methods and evaluate their applications on real-world domains, such as Remote Sensing, Bioinformatics and Medical field. SPONSORS GNCS-Gruppo Nazionale per il Calcolo Scientifico IEEE-Neural Networks Council INNS-International Neural Network Society SIREN-Italian Neural Networks Society University of Salerno,Italy DIRECTORS OF THE COURSE DIRECTORS OF THE SCHOOL Nathan Intrator (Israel and USA) Michael Jordan (USA) Francesco Masulli (Italy) Maria Marinaro (Italy) LECTURERS Leo Breiman, University of California at Berkeley, California, USA Lorenzo Bruzzone, University of Trento, Trento, Italy Thomas G. Dietterich, Oregon State University, Oregon, USA Cesare Furlanello, Istituto per la Ricerca Scientifica e Tecnologica, Trento, Italy Giuseppina C. Gini, Politecnico di Milano, Milano, Italy Tin Kam Ho, Bell Laboratories, New Jersey, USA Nathan Intrator, Tel Aviv Univ., Israel and Brown Univ., Providence, Rhode Island, USA Ludmila I. Kuncheva, University of Wales, Bangor, UK Francesco Masulli, University of Pisa, Italy Stefano Merler, Istituto per la Ricerca Scientifica e Tecnologica, Trento, Italy Fabio Roli, University of Cagliari, Cagliari, Italy Giorgio Valentini, University of Genova, Italy PLACE International Institute for Advanced Scientific Studies E.R. Caianiello (IIASS) Via Pellegrino 19, 84019 Vietri sul Mare, Salerno (Italy) POETIC TOUCH Vietri (from "Veteri", its ancient Roman name) sul Mare ("on sea") is located within walking distance from Salerno and marks the beginning of the Amalfi coast. Short rides take to Positano, Sorrento, Pompei, Herculaneum, Paestum, Vesuvius, or by boat, the islands of Capri, Ischia, and Procida. Velia (the ancient "Elea" of Zeno and Parmenide) is a hundred kilometers farther down along the coast. GENERALITIES Recently, driven by application needs, multiple classifier combinations have evolved into a practical and effective solution for real-world pattern recognition tasks. The idea appears in various disciplines (including Machine Learning, Neural Networks, Pattern Recognition, and Statistics) under several names: hybrid methods, combining decisions, multiple experts, mixture of experts, sensor fusion and many more. In some cases, the combination is motivated by the simple observation that classifier performance is not uniform across the input space and different classifiers excel in different regions. Under a Bayesian framework, integrating over expert distribution leads naturally to expert combination. The generalization capabilities of ensembles of learning machines have been interpreted in the framework of Statistical Learning Theory and in the related theory of Large Margin Classifiers. There are several ways to use more than one classifier in a classification problem. A first "averaging" approach consists of generating multiple hypotheses from a single or multiple learning algorithms, and combining them through majority voting or different linear and non linear combinations. A "feature-oriented" approach is based on different methods to build ensembles of learning machines by subdividing the input space (e.g., random subspace methods, multiple sensors fusion, feature transformation fusion). "Divide-and-conquer" approaches isolate the regions in input space on which each classifier performs well, and direct new input accordingly, or subdivide a complex learning problem in a set of simpler subproblems, recombining them using suitable decoding methods. A "sequential-resampling" approach builds multiple classifier systems using bootstrap methods in order to reduce variance (bagging) or jointly bias and unbiased variance (boosting). There are fundamental questions that need to be addressed for a practical use of this collection of approaches: What are the theoretical tools to interpret possibly in a unified framework this multiplicity of ensemble methods? What is gained and lost in a combination of experts, when is it preferable to alternative approaches? What types of data are best suitable to expert combination? What types of experts are best suited for combinations? What are optimal training methods for experts which are expected to participate in a collective decision? What combination strategies are best suited to a particular problem and to a particular distribution of the data? What are the statistical methods and the appropriate benchmark data to evaluate multiclassifier systems? The school will address some of the above questions from a theoretical and empirical view point and will teach students about this exciting and very promising field using current state of the art data sets for pattern recognition, classification and regression. The main goals of the school are: 1. Offer an overview of the main research issues of ensemble methods from the different and complementary perspectives of Machine Learning, Neural Networks, Statistics and Pattern Recognition. 2. Offer theoretical tools to analyze the diverse approaches, and critically evaluate their applications. 3. Offer practical and theoretical tools to develop new ensemble methods and analyze their application on real-world problems. FORMAT The meeting will follow the usual format of tutorials and panel discussions together with poster sessions for contributed papers. A demo lab with four Linux workstations will be available to the participants for testing and comparing ensemble methods. There will be a network of wireless 11MHz connection available so that students arriving with their laptops and an appropriate wireless communication card can stay connected while at the meeting area. DURATION Participants are expected to arrive in time for the evening meal on Sunday Sept 22th and depart on Saturday Sept 28th. Sessions will take place from Monday Sept 23th to Saturday Sept 28th. PROCEEDINGS The proceedings will be published in the form of a book containing tutorial chapters written by the lecturers and possibly shorter papers from other participants. One free copy of the book will be distributed to each participant after the school. LANGUAGE The official language of the school will be English. POSTER SUBMISSION There will be a poster session for contributed presentations from participants. Proposals consisting of a one page abstract for review by the organizers should be submitted with applications. REGISTRATION FEE Master and PhD Students: 650,00 Euro Academic Participants (govt/univ): 800,00 Euro Industrial Participants: 1.100,00 Euro The fee includes accommodation (3 stars hotel - double room), meals and a copy of the proceedings of the school. Transportation is not included. A supplement of 20 Euro per night should be paid for single room. Members of sponsoring organizations will receive a discount of 50 Euro on the registration fee. A few scholarships are available for students who are otherwise unable to participate at the school. Payment details will be notified with acceptance of applications. ELIGIBILITY The school is open to all suitably qualified scientists. People with few years of experience in the field should include a recommendation letter of their supervisor. APPLICATION PROCEDURE Important Dates: Application deadline: July 1, 2002 Notification of acceptance: July 10, 2002 Registration fee payment deadline: July 20, 2002 School Sept 22-28 2002 Places are limited to a maximum of 60 participants in addition to the lecturers. These will be allocated on a first come, first served basis. ********************************************************************** APPLICATION FORM Title: ............................................................... Family Name: ......................................................... Other Names:.......................................................... Mailing Address (include institution or company name if appropriate): ..................................................................... ..................................................................... ..................................................................... ..................................................................... ..................................................................... Phone:......................Fax:...................................... E-mail: .............................................................. Date of Arrival : .................................................... Date of Departure: ................................................... Are you sending the abstract of a poster? yes/no (delete the alternative which does not apply) Are you applying for a scholarship? yes/no (delete the alternative which does not apply) If yes please include a justification letter for the scholarship request. ***************************************************************** Please send the application form together the recommendation letter by electronic mail to: iiass.vietri at tin.it, subject: summer school; or by fax to: +39 089 761 189 (att.ne Prof. M. Marinaro) or by ordinary mail to the address below: IIASS Via Pellegrino 19, I-84019 Vietri sul Mare (Sa) Italy WEB PAGE OF THE COURSE The web page of the course is http://www.iiass.it/school2002 and will contain all the updates related to the course. At http://www.iiass.it/school2002/ensemble-lab.html a web portal to ENSEMBLE METHODS is in development including pointers to relevant papers, data-bases and software. Contributions to this portal are kindly requested to all researchers involved in this area. Please send all contributions to Giorgio Valentini (valenti at disi.unige.it). FOR FURTHER INFORMATION PLEASE CONTACT: Prof. Francesco Masulli DISI & INFM email: masulli at ge.infm.it Via Dodecaneso 35 fax: +39 010 353 6699 16146 Genova (Italy) tel: +39 010 353 6604 From chris at dcs.bbk.ac.uk Mon Jun 24 07:26:14 2002 From: chris at dcs.bbk.ac.uk (Chris Christodoulou) Date: Mon, 24 Jun 2002 12:26:14 +0100 (BST) Subject: Paper available: A Spiking Neuron Model: Applications and Learning Message-ID: Dear all, I would like to announce the availablility of the following paper that will appear in one of the forthcoming issues of Neural Networks. A preprint of the paper can be downloaded from: http://www.dcs.bbk.ac.uk/~chris/papers/nn_pprnt.pdf or http://www.tech.plym.ac.uk/soc/staff/GuidBugm/pub/nn2002.pdf .............................................................. Title: A Spiking Neuron Model: Applications and Learning Authors: Chris Christodoulou, Guido Bugmann and Trevor G. Clarkson (in press, Neural Networks) Abstract This paper presents a biologically-inspired, hardware-realisable spiking neuron model, which we call the Temporal Noisy-Leaky Integrator (TNLI). The dynamic applications of the model as well as its applications in Computational Neuroscience are demonstrated and a learning algorithm based on postsynaptic delays is proposed. The TNLI incorporates temporal dynamics at the neuron level by modelling both the temporal summation of dendritic postsynaptic currents which have controlled delay and duration and the decay of the somatic potential due to its membrane leak. Moreover, the TNLI models the stochastic neurotransmitter release by real neuron synapses (with probabilistic RAMs at each input) and the firing times including the refractory period and action potential repolarisation. The temporal features of the TNLI make it suitable for use in dynamic time-dependent tasks like its application as a motion and velocity detector system presented in this paper. This is done by modelling the experimental velocity selectivity curve of the motion sensitive H1 neuron of the visual system of the fly. This application of the TNLI indicates its potential applications in artificial vision systems for robots. It is also demonstrated that Hebbian-based learning can be applied in the TNLI for postsynaptic delay training based on coincidence detection, in such a way that an arbitrary temporal pattern can be detected and recognised. The paper also demonstrates that the TNLI can be used to control the firing variability through inhibition; with 80% inhibition to concurrent excitation, firing at high rates is nearly consistent with a Poisson-type firing variability observed in cortical neurons. It is also shown with the TNLI, that the gain of the neuron (slope of its transfer function) can be controlled by the balance between inhibition and excitation, the gain being a decreasing function of the proportion of inhibitory inputs. Finally, in the case of perfect balance between inhibition and excitation, i.e., where the average input current is zero, the neuron can still fire as a result of membrane potential fluctuations. The firing rate is then determined by the average input firing rate. Overall this work illustrates how a hardware-realisable neuron model can capitalise on the unique computational capabilities of biological neurons. Keywords: Spiking Neuron Model; Temporal Noisy-Leaky Integrator; Motion detection; Directional selectivity; postsynaptic delay learning; temporal pattern detection; high firing variability; inhibition. * * * Dr Chris Christodoulou chris at dcs.bbk.ac.uk Chris.Christodoulou at kcl.ac.uk School of Computer Science and Information Systems Birkbeck College, University of London Malet Street, London WC1E 7HX, UK Tel. (+44) 20-7631 6718, Fax (+44) 20-7631 6727 From giro-ci0 at wpmail.paisley.ac.uk Wed Jun 26 09:11:52 2002 From: giro-ci0 at wpmail.paisley.ac.uk (Mark Girolami) Date: Wed, 26 Jun 2002 14:11:52 +0100 Subject: Lecturing Positions Available Message-ID: Dear Connectionists, There are two academic (faculty) positions now available at the University of Paisley and I would be very pleased to hear from subscribers to Connectionists who are interested in these posts. Rgds Mark ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Lecturers in Computing (2)School of Information & Communication TechnologiesFaculty of Communications, Engineering & Science Salary up to 32,537 Two lecturing positions are now available one post will be based at the Paisley Campus of the University and the other will be based at Crichton Campus in Dumfries. You must have at least an upper second class honours degree or equivalent qualification in Computer Science or a related area, a commitment to teaching to a high standard and a good record of research activity. A PhD. Degree is desirable. You will also be expected to participate in administration and to establish research activities in an appropriate area. Informal enquiries to Professor Mark Girolami on 0141 848 3317 E-mail: mark.girolami at paisley.ac.uk. Recruitment packs may be obtained from the Department of Human Resources, University of Paisley, PAISLEY, PA1 2BE (Tel 0141 848 3692 - answering service outwith normal working hours). Closing date: 5 July 2002. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Professor. M.A Girolami PhD Associate Head of School and Chair of Applied Computational Intelligence School of Information and Communication Technologies University of Paisley High Street Paisley, PA1 2BE Tel: +44 (0)141 848 3317 Fax +44 (0)141 848 3542 http://cis.paisley.ac.uk/giro-ci0 Legal disclaimer -------------------------- The information transmitted is the property of the University of Paisley and is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Statements and opinions expressed in this e-mail may not represent those of the company. Any review, retransmission, dissemination and other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender immediately and delete the material from any computer. -------------------------- From mayank at MIT.EDU Wed Jun 26 18:00:53 2002 From: mayank at MIT.EDU (Mayank R Mehta) Date: Wed, 26 Jun 2002 18:00:53 -0400 (EDT) Subject: Paper Available: Role of Experience & Oscillations in Transfomring a Rate Code into a Temporal Code In-Reply-To: <15414.2114.190081.556538@coltrane.ucsf.edu> Message-ID: The following paper can be downloaded from my home page http://www.mit.edu/~mayank This is the final draft of a paper that has appeared in Nature, 417, 741-746 (2002). Summary: While precise spike timing is critical for synaptic plasticity, neither the mechanism for generating a temporal code that establishes such precise spike timing in vivo, nor the effect of experience on such a temporal code have been established. Here we propose a mechanism by which a temporal code can be generated through an interaction between an asymmetric rate code and oscillatory inhibition. Consistent with the predictions of our model the rate and temporal codes of hippocampal pyramidal neurons are highly correlated. Furthermore, the temporal code becomes more robust with experience. The resulting spike timing satisfies the temporal order constraints of Hebbian learning. Thus, oscillations and receptive field asymmetry may play a critical role in temporal sequence learning. ---------------- Apologies for cross-postings. Cheers! -Mayank -----------------------+----------------------------+ Mayank R. Mehta | Email: Mayank at MIT.edu | E18-366, M.I.T. | Phone: 617 252 1841 | 50 Ames St. | FAX: 617 452 4120 | Cambridge, MA 02139 | http://www.mit.edu/~mayank | -----------------------+----------------------------+ From Frederic.Alexandre at loria.fr Fri Jun 28 04:36:01 2002 From: Frederic.Alexandre at loria.fr (Frederic Alexandre) Date: Fri, 28 Jun 2002 10:36:01 +0200 Subject: Two PhD studenship in computer science Message-ID: <3D1C1FF1.DC10560A@loria.fr> The CORTEX team at LORIA/INRIA Lorraine in Nancy, FRANCE is looking for 2 PhD students. http://www.loria.fr/LORIA/LORIA-test/EQUIPES/index.en.php?equipe=Cortex Applications (CV and accompanying letter) are invited for two fully funded Phd studenship, one in the area of neuroscience and modelling, the other in the area of knowledge extraction using neural network. PROJECT 1: In the framework of the MirrorBot European project, between neurosciences and computer science, we aim at designing a biological neural model for the control of an autonomous robot. PROJECT 2: In the framework of an industrial project using extensive geographical data, we aim at designing a neural network for automatic knowledge extraction Details below --- PROJECT 1 In tight connection with CORTEX research themes and taking benefit from our experience in the biologically neural modelling area, we aim at designing a neural networks for the control of an autonomous robot (PeopleBot) in a real but simplified unknown environment. Throughout the many interactions between the robot and the world (objects, positions, colors), we want to study the emergence of semantic during learning. This requires the conception of motor and perceptive systems as well as a sensori-motor coordination system. Moreover, we would like to emphasize the need for different memory systems like declarative memory, procedural memory and working memory. This work will take place in interaction with members of the team and neurobiologists participating to this project. COMPETENCE: The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science or other related area with knowledge and interest in neural networks and modelling. Fluent English (or French) is required. Programming skills (eg C/C++) are desired. CONTACT: Dr. Nicolas Rougier Nicolas.Rougier at loria.fr LORIA laboratory Nancy, France APPLICATION: Closing date for applications: July 15, 2002. --- PROJECT 2 Today, the problem is no longer the need for information but the way to focus on the relevant one and to be able to extract the implicit knowledge contained within. From one hand, internal structures of neural networks are responsible for their good performances in classification and data foraging and in the other hand these internal structures have been hardly studied in terms of information representation and implicit knowledge. Consequently, this work aims at a better understanding of these internal structures and implicit knowledge but also at designing a working model for knowledge extraction. This work will take place in collaboration with the national geological research company (BRGM) in charge of natural ressources in France. COMPETENCE: The successful candidate should have an MSc grade (or about to get one) or equivalent in computer science or other related area with knowledge and interest in neural networks. Fluent English (or French) is required. Programming skills (eg Java, C/C++) and knowledge in AI and statistics are desired. CONTACT: Dr. Laurent Bougrain Laurent.Bougrain at loria.fr LORIA laboratory Nancy, France APPLICATION: Closing date for applications: July 15, 2002. From sue at soc.plym.ac.uk Sun Jun 30 07:34:09 2002 From: sue at soc.plym.ac.uk (Sue Denham) Date: Sun, 30 Jun 2002 12:34:09 +0100 Subject: Research positions in visual and auditory modelling and psychophysics Message-ID: <00b201c2202a$0d77b760$6a4da38d@sue1> Plymouth Institute of Neuroscience Applications are invited for the following positions: 2 postdoctoral research fellowships and 2 research studentships in visual and auditory modelling and psychophysics. The open positions are part of a project, "Attend-to-learn and learn-to-attend with neuromorphic VLSI", funded by the European Community (IST-2001-38099), which combines perceptual, computational, and hardware research. In addition to the University of Plymouth, UK, the project also involves, ETH Zurich and University of Bern, Switzerland, the National Institute of Health, Rome, Italy, Siemens AG, Germany, and the UC Davis, USA. Research at the Plymouth Institute of Neuroscience received a rating of 5 in the 2001 RAE exercise. The aim of the project is to develop a general architecture for memory-guided attention using both software modelling and neuromorphic VLSI technology. The architecture will be tested on natural visual and auditory stimuli and its performance compared to human observers/listeners. Plymouth will be responsible for developing visual and auditory stimuli that are restricted enough to be tractable by an artificial system, yet rich enough for psychophysical tests of attention and learning in human observers/listeners. Specific tasks will include: a.. designing feature spaces that efficiently encode of natural images and sounds b.. conducting psychophysical experiments to characterise attention and learning with novel classes of synthetic images and sounds c.. developing computational models for visual and auditory feature saliency and feature tracking d.. providing performance benchmarks that are suitable for testing the models and hardware components developed by project partners e.. disseminating the results at international conferences. Applicants for the postdoctoral research fellowships should have, or expect to obtain, a Ph.D. in Psychophysics, Computational Modelling, or a related field. Experience in information theoretic analysis is also desirable. Applicants for research studentships should have, or expect to obtain, a good honours degree in Psychology, Neuroscience, Physics, Computing, or related fields. The ideal candidate will possess good analytical, experimental and computational skills. All posts are available from 1 September 2002 for 3 years. Salaries will be internationally competitive and reflect the successful candidate's qualifications. For further information, please contact Prof. Jochen Braun (visual psychophysics and modelling, +44 1752 232 711, achim at pion.ac.uk) or Dr. Sue Denham (auditory psychophysics and modelling, +44 1752 232610, sue at pion.ac.uk).