From dandre at BodyMedia.Com Mon Feb 2 14:36:19 2004 From: dandre at BodyMedia.Com (Dave Andre) Date: Mon, 2 Feb 2004 14:36:19 -0500 Subject: ICML competition and workshop on Physiological Data Modeling Message-ID: <4AA3DF1F6B40B045AC9D384A31BB7FD6126C43@mailserver.private.bodymedia.com> ICML 2004 workshop The Physiological Data Modeling Contest (PDMC) A Machine Learning Challenge July 8th, 2004 Banff, Alberta, Canada http://www.cs.utexas.edu/users/sherstov/pdmc Physiological data offers many challenges to the machine learning community including dealing with large amounts of data, sequential data, issues of sensor fusion, and a rich domain complete with noise, hidden variables, and significant effects of context. The Physiological Data Modeling Contest (PDMC) will challenge participants to exhibit machine learning algorithms for classification and regression on data taken using BodyMedia (www.bodymedia.com ) wearable body monitors. These devices collect and store continuous data from multiple sensors packaged in an unobtrusive armband. To enable this contest, BodyMedia is providing a rich dataset comprising several months of data from more than two dozen subjects. Features include raw data such as skin temperature, heat flux, galvanic skin response, and vital statistics of the user. The tasks include detecting classes of activity being engaged in by the user (e.g. walking, sleeping, watching TV, etc.) based on the values of these features. The training data is available now; the test data will be released approximately four weeks before the workshop; and the competitors will submit entries consisting of predictions for this test data shortly before ICML 2004. At the workshop, the performance of each entrant will be announced and the various techniques discussed and analyzed in detail. A small prize will be awarded to the winning competitors. Entrants will be invited to contribute papers describing their approaches to a special issue of a technical journal or magazine. The intended audience is machine learning researchers and professors currently teaching machine learning classes. Complete details are available on the contest website: http://www.cs.utexas.edu/users/sherstov/pdmc Important dates (tentative) =============== NOW: Send your expression of interest to the organizers 2 April: Notice of intent to participate due 1 June : Test data set released and announced 18 June : Entry submission deadline 8 July : Workshop/results Organizers ========== David Andre, dandre at BodyMedia.com, BodyMedia, Inc. (co-chair) Peter Stone, pstone at cs.utexas.edu, UT Austin (co-chair) Alexander Sherstov, sherstov at cs.utexas.edu, UT Austin Maxence Crossley, max at BodyMedia.com, BodyMedia, Inc. From Pierre.Bessiere at imag.fr Mon Feb 2 07:59:18 2004 From: Pierre.Bessiere at imag.fr (bessiere) Date: Mon, 02 Feb 2004 13:59:18 +0100 Subject: Bayesian Robot Programming Message-ID: We are pleased to announce the publication of the following paper in the January 2004 issue of "Autonomous Robots" a Kluewer publication: Bayesian Robot Programming ************************** Freely available at the following URL: http://www.kluweronline.com/issn/0929-5593/contents Abstract: --------- We propose a new method to program robots based on Bayesian inference and learning. It is called BRP for Bayesian Robot Programming. The capacities of this programming method are demonstrated through a succession of increasingly complex experiments. Starting from the learning of simple reactive behaviors, we present instances of behavior combination, sensor fusion, hierarchical behavior composition, situation recognition and temporal sequencing. This series of experiments comprises the steps in the incremental development of a complex robot program. The advantages and drawbacks of BRP are discussed along with these different experiments and summed up as a conclusion. These different robotics programs may be seen as an illustration of probabilistic programming applicable whenever one must deal with problems based on uncertain or incomplete knowledge. The scope of possible applications is obviously much broader than robotics. Keywords: --------- Bayesian robot programming, control of autonomous robots, computational architecture for autonomous systems, theory of autonomous systems Authors: -------- Olivier Lebeltel, Pierre Bessi?re, Julien Diard & Emmanuel Mazer ___________________________________________________________________ Dr Pierre BESSIERE ? CNRS ************************ Laboratoire GRAVIR ? Institut IMAG INRIA ? Rh?ne-Alpes 655 avenue de l?Europe 38034 St Ismier ? FRANCE Work?: +33 (0) 4.76.61.55.09 Fax?: +33 (0) 4.76.61.52.10 Mailto:Pierre.Bessiere at imag.fr WWW?: http://www-laplace.imag.fr ___________________________________________________________________ From clabiouse at ulg.ac.be Mon Feb 2 10:48:17 2004 From: clabiouse at ulg.ac.be (Christophe Labiouse) Date: Mon, 2 Feb 2004 16:48:17 +0100 Subject: LAST CALL OF PAPERS: Small Group Meeting on Social Connectionism (16-19 June 2004, Brussels, Belgium) Message-ID: <003c01c3e9a3$fa35d980$0b3ba58b@clabiouse> Small Group Meeting on Social Connectionism=20 16-19 June 2004, Brussels, Belgium [Van Overwalle (Vrije Universiteit Brussel, Belgium) and Christophe Labiouse (Universit=E9 de Li=E8ge, Belgian National Fund of Scientific Research, Belgium)] The deadline for submitting abstracts is Monday 16th February 2004. The abstract should include a list of authors, including the first author's correspondence address and email, and a summary of not more than 150 words, and should be sent to Frank Van Overwalle (Frank.VanOverwalle at vub.ac.be). =20 A number of renowned researchers in the field of connectionism will present a state-of-the-art presentation of current and novel developments in their field: Bob French (topic: dual-memory models) and Axel Cleeremans (topic: implicit learning). Moreover, renowned social researchers who worked on connectionist models in social cognition have confirmed to come to the meeting (Stephen Read and Yoshi Kashima). The meeting is limited to 20-25 participants (also PhD students or postgraduates) with an interest in social connectionism. We expect people working in the fields such as covariation, causal judgment, illusory correlation, cognitive dissonance, person perception, stereotyping, attitudes and other judgmental processes. We would also like to invite cognitive researchers or neuropsychologists working in the connectionist or associative learning domain. Topic of the meeting In the last decade, we have seen an upsurge of articles attempting to model social processes within a connectionist or computational framework. The aim of this small group meeting is to gain further insight in the possible connectionist processes that underlie social thinking by comparing and testing these different connectionist approaches. Papers are invited that propose improved models or that evaluate and discuss the merits of existing models. What are the basic properties of proposed connectionist architectures? The amount of known and novel findings in the field that it can account for? The amount of new hypothesis it generates? The breadth of empirical support? Apart from discussing the merits of specific connectionist implementations, we also want to address broader questions. What are general weaknesses of existing models? What are novel directions and phenomena that these models should attempt to cover? For instance, how are motivation, attention, and goal-directed behavior accounted for? How should we make a distinction between episodic and semantic memory? Between implicit and explicit learning and reasoning? What is the neuropsychological basis of the model's core mechanisms? Answers to these questions might improve our understanding of the connectionist underpinnings of social reasoning. Frank Van Overwalle Vrije Universiteit Brussel Department of Psychology tel: +32 2 629 25 18 Pleinlaan 2 fax: +32 2 629 24 89 B - 1050 Brussel e-mail: Frank.VanOverwalle at vub.ac.be Belgium www.vub.ac.be/PESP/VanOverwalle.html From jianfeng at sussex.ac.uk Tue Feb 3 16:23:25 2004 From: jianfeng at sussex.ac.uk (jianfeng@sussex.ac.uk) Date: Tue, 03 Feb 2004 21:23:25 +0000 Subject: postdocs In-Reply-To: References: Message-ID: <1075843405.4020114d1f7b6@webmail.sussex.ac.uk> The Babraham Institute is an independent charity whose research is applied to find new medicines and treatments for human diseases. Biological Statistician (Ref. KK/BS/ST) Starting salary range 26,000 - 33,500 p.a. A statistician with experience of multivariate analysis is required to assist in developing appropriate analytical protocols and visual representation of data from multiple cell recordings in brain sensory systems. This will also involve collaboration with a Neuroinformatics group at the University of Sussex. It is expected that the individual will subsequently apply these skills to other complex biological data and provide more general statistical advice and training for Institute scientists. The successful candidate should have a degree in biology, computing, statistics or mathematics and a higher degree in statistics or mathematics. For informal enquiries please contact Professor Keith Kendrick (keith.kendrick at bbsrc.ac.uk) Bioinformatician (Ref. WR/BIO/ST) Starting salary range 21,700 - 25,200 p.a. The bioinformatics post is at a junior level and will assist existing group members in the Bioinformatics facility with the provision of support, training and development for the research groups at the Institute. An initial focus will be for a project investigating evolutionary conservation of imprinted genes (in collaboration with the Sanger Institute). This position would suit someone wishing to pursue more formally their existing interest in bioinformatics. The successful candidate should have a degree in a Biological science. For informal enquiries please contact Dr Wolf Reik (wolf.reik at bbsrc.ac.uk) It is anticipated that funding will be available for 3 years initially for both positions. Starting salary for both posts will be according to qualifications and experience. Benefits include pension scheme. Further details and application forms available from the Personnel Office, The Babraham Institute, Babraham, Cambridge CB2 4AT. Tel. 01223 496000, e-mail: babraham.personnel at bbsrc.ac.uk. Closing date for completed applications is 27th February 2004. Please quote the appropriate reference number when applying. AN EQUAL OPPORTUNITIES EMPLOYER An Institute supported by the Biotechnology and Biological Sciences Research Council From terry at salk.edu Tue Feb 3 13:04:53 2004 From: terry at salk.edu (Terry Sejnowski) Date: Tue, 3 Feb 2004 10:04:53 -0800 (PST) Subject: NEURAL COMPUTATION 16:3 In-Reply-To: <200401080215.i082FOu12798@purkinje.salk.edu> Message-ID: <200402031804.i13I4rE81136@purkinje.salk.edu> Neural Computation - Contents - Volume 16, Number 3 - March 1, 2004 ARTICLE Geometrical Computations Explain Projection Patterns of Long-Range Horizontal Connections in Visual Cortex Ohad Ben-Shahar and Steven Zucker NOTES Mean Instantaneous Firing Frequency is Always Higher than the Firing Rate Petr Lansky, Roger Rodriguez and Laura Sacerdote Kalman Filter Control Embeded into the Reinforcement Learning Framework Istvan Szita and Andras Lorincz LETTERS Rapid Processing and Unsupervised Learning in a Model of the Cortical Macrocolumn Jorg Luecke and Christoph von der Malsburg Temporally Asymmetric Learning Supports Sequence Processing in Multi-Winner Self-Organizing Maps Reiner Schultz and James A. Reggia Neuronal Bases of Perceptual Learning Revealed by a Synaptic Balance Scheme Osamu Hoshino How the Shape of Pre-and Postsynaptic Signals Can Influence STDP: A Biophysical Model Ausra Saudargiene, Bernd Porr and Florentin Worgotter Self-Organizing Dual Coding Based on Spike-Time-Dependent Plasticity Naoki Masuda and Kazuyuki Aihara ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2004 - VOLUME 16 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $108 $54 $57.78 Individual $95 $101.65 $143 $85 $90.95 Institution $635 $679.45 $689 $572 $612.04 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From ahu at cs.stir.ac.uk Wed Feb 4 18:28:05 2004 From: ahu at cs.stir.ac.uk (Dr. Amir Hussain) Date: Wed, 4 Feb 2004 23:28:05 -0000 Subject: Final CFP: BICS'2004: Upcoming (Extended) Deadline 29 February 2004 Message-ID: <000101c3eb76$9a0f0610$1f95fc3e@DrAmir> Brain Inspired Cognitive Systems (BICS'2004): August 29 - September 1, 2004, University of Stirling, Scotland Final Call for Papers: UPCOMING (EXTENDED) PAPER SUBMISSION DEADLINE February 29, 2004 NEW: Authors of selected papers will be invited to submit an extended version of their paper for publication in a special issue of the Neurocomputing Journal, published by Elsevier Science B.V. (http://www.elsevier.nl/locate/neucom) Why This Conference, and who should Attend: The science of neural computation focuses on mathematical aspects for solving complex practical problems. It also seeks to help neurology, brain theory and cognitive psychology in the understanding of the functioning of the nervous system by means of computational models of neurons, neural nets and sub-cellular processes. BICS2004 aims to become a major point of contact for research scientists, engineers and practitioners throughout the world in the fields of cognitive and computational systems inspired by the brain and biology. Participants will share the latest research, developments and ideas in the wide arena of disciplines encompassed under the heading of BICS2004: First International ICSC Symposium on Cognitive Neuro Science (CNS 2004) (from computationally inspired models to brain-inspired computation) Chair: Prof. Igor Aleksander, Imperial College London, U.K Second International ICSC Symposium on Biologically Inspired Systems (BIS 2004) Chair: Prof. Leslie Smith, University of Stirling, U.K. Third International ICSC Symposium on Neural Computation (NC'2004) Chair: Dr. Amir Hussain, University of Stirling, U.K. http://www.icsc-naiso.org/conferences/bics2004/bics-cfp.html - Dr. Amir Hussain Dept. of Computing Science University of Stirling Stirling FK9 4LA, UK Email: ahu at cs.stir.ac.uk http://www.cs.stir.ac.uk/~ahu Tel/Fax: (+44) 01786 - 467437 / 464551 -- From bernhard.schoelkopf at tuebingen.mpg.de Thu Feb 5 03:45:28 2004 From: bernhard.schoelkopf at tuebingen.mpg.de (Bernhard Schoelkopf) Date: Thu, 5 Feb 2004 09:45:28 +0100 Subject: job openings in Tuebingen: group leaders, PhD student Message-ID: <015201c3ebc4$686458b0$ff85260a@ebro> Dear Connectionists, 1. There is a job opening for three research group leaders (C3/W2, corresponding to Associate Professor) in the Miescher Lab of the Max Planck Society, broad enough to cover the whole spectrum of the connectionists list. The Miescher Lab is on our campus, so there would be ample opportunity to interact. For details, see http://www.kyb.tuebingen.mpg.de/misc/jobs/pics/GablerMPI-Independent.pdf From ted.carnevale at yale.edu Thu Feb 5 10:02:33 2004 From: ted.carnevale at yale.edu (Ted Carnevale) Date: Thu, 05 Feb 2004 10:02:33 -0500 Subject: 2004 NEURON Simulator Meeting Message-ID: <40225B09.4040506@yale.edu> What: the first NEURON Simulator Meeting When: 9 AM Friday - Noon Sunday, May 14-16, 2004 Where: Northwestern University, Chicago, IL Registration is now open for the first NEURON Simulator Meeting. The purpose of this meeting is to bring together people who are interested in using computational modeling in neuroscience research and education--especially current users of NEURON, and others who are interested in using it--in order to: * share knowledge about technical aspects of computational modeling * keep informed about the latest advances and "best practices" in the use of NEURON * encourage participation in NEURON's development * stimulate communication and collaboration This conference will alternate between single track sessions for presentations and discussions on topics of general interest, and multiple track breakout sessions for smaller group interactions. We invite proposals for talks, symposia, tutorials, and workshops (see "Proposals Invited" below). Registration is limited to 50 individuals on a first-come, first serve basis. For more information see http://www.neuron.yale.edu/neuron/nsm2004/nsm2004.html Proposals Invited We invite participants to propose and be responsible for talks, symposia, tutorials, or workshops. Each individual whose proposal is accepted will be eligible for reimbursement of a substantial amount of travel costs to attend the meeting. See http://www.neuron.yale.edu/neuron/nsm2004/nsm2004.html for more information. --Ted From bernhard.schoelkopf at tuebingen.mpg.de Thu Feb 5 03:45:28 2004 From: bernhard.schoelkopf at tuebingen.mpg.de (Bernhard Schoelkopf) Date: Thu, 5 Feb 2004 09:45:28 +0100 Subject: job openings in Tuebingen: group leaders, PhD student Message-ID: <015201c3ebc4$686458b0$ff85260a@ebro> Dear Connectionists, 1. There is a job opening for three research group leaders (C3/W2, corresponding to Associate Professor) in the Miescher Lab of the Max Planck Society, broad enough to cover the whole spectrum of the connectionists list. The Miescher Lab is on our campus, so there would be ample opportunity to interact. For details, see http://www.kyb.tuebingen.mpg.de/misc/jobs/pics/GablerMPI-Independent.pdf 2. Our department "Empirical Inference for Machine Learning and Perception" is looking for a PhD student candidate. Applicants should be academically outstanding and highly motivated to do basic research. For more information on us, see http://www.kyb.tuebingen.mpg.de/bs. The starting date is flexible. best regards Bernhard From neil at dcs.shef.ac.uk Fri Feb 6 06:13:20 2004 From: neil at dcs.shef.ac.uk (Neil Lawrence) Date: Fri, 6 Feb 2004 11:13:20 -0000 Subject: PhD Studentship in Machine Learning and Speech Message-ID: <004e01c3eca2$3aa58910$e108a78f@windows.dcs.shef.ac.uk> Dear Connectionists, We would like to announce the following PhD studentship sponsored by the University of Sheffield. "Learning to Recognise Speech in Everyday Noisy Environments" The student will be jointly supervised by Prof. Martin Cooke and Dr Neil Lawrence in the University of Sheffield's Department of Computer Science. The closing date for applications is 5th March 2004 For more details on the project and how to apply see http://www.dcs.shef.ac.uk/~neil/studentship.html. Regards, Martin Cooke Neil Lawrence From Michael.Berthold at uni-konstanz.de Fri Feb 6 03:51:10 2004 From: Michael.Berthold at uni-konstanz.de (Michael Berthold) Date: Fri, 6 Feb 2004 09:51:10 +0100 Subject: Scholarships for PhD Program at Konstanz University Message-ID: <20040206085108.B755B1F800A@viribus.rz.uni-konstanz.de> The Department of Computer and Information Science at Konstanz University, Germany is please to announce its new PhD Graduate Program. The program is supported by the Deutsche Forschungsgemeinschaft (DFG) and offers various scholarships for graduate students and one postdoc. The topics will mainly focus around the theme of the graduate school: "Explorative Analysis and Visualization of Large Information Spaces" Please advise all interested students in your department of this great opportunity. The first deadline for applications is March 15, 2004. Detailed information about the program can be found at: http://www.inf.uni-konstanz.de/gk -- Michael R. Berthold, Professor phone: +49 (07531) 88-2202 ALTANA-Chair for Applied Computer Science fax: +49 (07531) 88-5132 (Bioinformatics and Information Mining) berthold at inf.uni-konstanz.de University of Konstanz, Germany www.inf.uni-konstanz.de/~berthold From bogus@does.not.exist.com Fri Feb 6 15:06:40 2004 From: bogus@does.not.exist.com () Date: 06 Feb 2004 15:06:40 -0500 Subject: No subject Message-ID: <1076098000.11288.218.camel@dijkstra.cs.wlu.edu> This CFP is also available in MSWord and PDF format, from: http://www.cs.wlu.edu/~levy/aaai04/AAAI_FSS04_CCCS_CFP.doc http://www.cs.wlu.edu/~levy/aaai04/AAAI_FSS04_CCCS_CFP.pdf Compositional Connectionism in Cognitive Modeling (AAAI 2004 Fall Symposium Series) October 22-24, 2004 Washington, D.C. U.S.A. This symposium will bring together connectionist and non-connectionist researchers to discuss and debate a topic of central concern in AI and cognitive science: the nature of compositionality. The open-ended productivity of the human capabilities aspired to by AI (e.g., perception, cognition, and language) is generally taken to be a consequence of compositionality; i.e., the ability to combine constituents recursively. The aim of this symposium is to expose connectionist researchers to the broadest possible range of conceptions of composition including those conceptions that pose the greatest challenge for connectionism while simultaneously alerting other AI and cognitive science researchers to the range of possibilities for connectionist implementation of composition. We therefore welcome and encourage submissions from both proponents and critics of connectionist representations, so long as the work described focuses on compositionality in an AI or in the modeling of cognition. Specifically, the symposium will seek to address the following issues: (1) What do we mean by compositional in terms of the relationship between the constituents and the composite? (2) What do we mean by the process of composition? (3) Are composition and decomposition (of a given type) completely symmetric operations? (4) What constituents (that are relevant to human agency) can be composed? (5) What are the consequences of having multiple interacting compositional systems? (6) To what extent do "structure-in-time" connectionist models, like Elman's Simple Recurrent Networks, represent compositional structure? (7) Do different connectionist models implement different aspects of compositionality, and can they be used in conjunction to address different aspects of composition? Plenary Speakers Dr. Jeffrey Elman, Dept. of Cognitive Science, University of California at San Diego Dr. Mark Steedman, School of Informatics, University of Edinburgh Submissions Standard talks at the symposium will be allocated 20 minutes for presentation, and papers will be limited to four pages. As the participants are expected to come from a wide range of disciplines, authors are asked to consider very carefully these constraints of brevity and variable background knowledge in planning their papers. Such constraints will be a primary criterion for evaluating submissions. Send submissions (in PDF, Postscript, or Microsoft Word format) to levys at wlu.edu (Simon D. Levy). Important Deadlines for Authors May 3, 2004 Submission due to organizers May 24, 2004 Notifications of acceptance sent by organizers August 31, 2004 Accepted camera-ready copy due to AAAI Registration The symposium will be limited to 40-60 participants. Those interested in attending should submit a paper (details above) or, if not intending to speak, submit a brief position paper or research abstract. Both sorts of submission (full paper or position / abstract) are due May 3, 2004. The symposium is intended for active participants, however there may be a limited number of places for interested individuals. These will be available on a first-come, first-served basis after the opening of general registration. General registration information will be available on the AAAI web site in July 2004. More detailed registration information will be available later and a separate Call for Participation will be made. Organizing Committee Simon D. Levy Computer Science Dept., Washington & Lee University Ross Gayler La Trobe University / Baycorp Advantage Value Solutions Pentti Kanerva Redwood Neuroscience Institute Chris Eliasmith Depts. of Philosophy / Systems Design Engineering, U. Waterloo Further Information For further information on the AAAI Symposia please see: http://www.aaai.org/Symposia/symposia.html http://www.aaai.org/Symposia/Fall/2004/fss-04.html -- Simon D. Levy Assistant Professor Computer Science Department Washington & Lee University Lexington, VA 24450 540-458-8419 (voice) 540-458-8479 (fax) levys at wlu.edu http://www.cs.wlu.edu/~levy From s.nolfi at istc.cnr.it Fri Feb 6 09:06:26 2004 From: s.nolfi at istc.cnr.it (Stefano Nolfi) Date: Fri, 6 Feb 2004 15:06:26 +0100 Subject: Special issue on Evolutionary Robotics Message-ID: <00e301c3ecba$68f2ca40$91419296@NOLFI> Call for Papers Special Issue on "Evolutionary Robotics: Looking forward" Connection Science journal Guest Editor : Stefano Nolfi Over the last ten years, or more, Evolutionary Robotics has attracted the interest of a large community of researchers with different research interests and backgrounds, ranging from AI and robotics, over biology and cognitive science, to the study of social behavior. Continuous progress in evolutionary robotics has led to a substantial maturation of the field and a clearer understanding of its potential and of its current limitations. The goal of this special issue is to encourage researchers working in this field not only to present their most recent research, but also to consider its relationship to other research fields, and to identify promising research directions for the future. Papers are welcome addressing (but not limited to) the following topics: Body and brain co-evolution, new materials Collective robotics and self-assembling Incremental and competitive evolution Open-ended evolutionary processes Evolution and evolvability Evolution, development and learning Representation and behavior Communication and language Cognitive capacities Specifically, the submitted papers are expected to: (a) focus on one or more related research topics, (b) explain the importance of the topic, the open problems and the different approaches discussed in the literature, (c) discuss the implications with respect to related research fields, (d) discuss the advantages and drawbacks of the evolutionary robotics approach with respect to other methodologies. Submission Instructions All manuscripts should be emailed to the guest editor (Stefano Nolfi, s.nolfi at istc.cnr.it) by May 15, 2004. Instructions for authors are available from: http://www.tandf.co.uk/journals/authors/ccosauth.asp. Manuscripts can either be full papers or shorter research notes (up to 4000 words). Important dates: Deadline for paper submission: May 15, 2004 Notification of acceptance: August 1, 2004 Final manuscript due: September 1, 2004 Publication of the special issue: December, 2004 Guest Editor: Stefano Nolfi Institute of Cognitive Sciences and Technology, CNR Viale Marx, 15 - 00137 - Rome - Italy e-mail: s.nolfi at istc.cnr.it voice: 0039-06-86090231 fax: 0039-06-824737 About the Journal Connection Science is an interdisciplinary scientific journal with a focus on the mechanisms of adaptation, cognition and intelligent behaviour in both living and artificial systems. The traditional scope of the journal has been broadened from connectionist research and neural computing to encompass work on other adaptive methods (e.g. evolutionary computing) as well as biologically inspired techniques and algorithms in applied domains. Papers should be related either to the behaviour of humans and/or other animals or to the underlying mechanisms. Papers submitted to the journal may be practical implementations, theoretical research or philosophical discussions. The submission of robotics research papers on issues raised by the interaction of agents with the environment or with other agents is particularly encouraged. ---------------------------------------------------------------------------- Stefano Nolfi Institute of Cognitive Sciences and Technologies, CNR Viale Marx, 15 - 00137 - Rome - Italy voice: 0039-06-86090231 fax: 0039-06-824737 e-mail:s.nolfi at istc.cnr.it www: http://gral.ip.rm.cnr.it/nolfi From ebaum at fastmail.fm Sun Feb 8 10:03:34 2004 From: ebaum at fastmail.fm (Eric Baum) Date: Sun, 8 Feb 2004 10:03:34 -0500 Subject: Hayek, Learning Go, Thought? Message-ID: <16422.20422.525811.937597@localhost.localdomain> My research website is back up at a new location, revised and with new material: http://www.whatisthought.com/eric.html Included are the papers on the Hayek model of evolutionary computing, including a new one applying the model to focussed web crawling, game playing papers, including new slides on A Learning Approach to Go, and DNA computing papers. If you were linked to my old NEC website, please revise your links. Also check out http://whatisthought.com , a website for the eponymous book, with ongoing posts of new research related to thought. From dummy at dsi.unifi.it Mon Feb 9 03:31:28 2004 From: dummy at dsi.unifi.it (Machine Learning Group, Univ. Firenze) Date: Mon, 09 Feb 2004 09:31:28 +0100 Subject: Postoctoral position in machine learning Message-ID: <5AEEEADE-5ADA-11D8-ADA2-000393A37AB8@dsi.unifi.it> The Neural Networks and Machine Learning group at Universita' di Firenze (Italy) is seeking one postdoctoral researcher to join an EU funded collaborative research project focused on algorithms and architectures for learning with probabilistic logic representations. The position will be funded on the three-year project APRIL II, in the 6th framework programme (Future and Emerging Technologies). APRIL II aims at developing a theoretical framework for probabilistic logic learning and demonstrate its effectiveness on significant applications, especially in the area of bioinformatics. Program partners are Freiburg University, Germany, Imperial College London, UK, Univ. of Helsinki, Finland, INRIA Rocquencourt, France, Universita' di Firenze, Italy, Aalborg University, Denmark, and Univ. of York, UK. Tight scientific collaboration is expected among all the research partners. We are especially interested in applicants with a strong background and experience in computational logic and machine learning. Experience in other related areas such as relational learning systems, probabilistic relational models, kernel methods on discrete structures, is also desiderable. Salary will be commensurate with qualifications and experience. The position is available immediately and can be renewed for up to three years. Researchers who have recently completed their Ph.D. few months are particularly encouraged to apply. Applicants are invited to send a CV and two letters of recommendation to Prof. Paolo Frasconi Dipartimento di Sistemi e Informatica Universit? di Firenze Via di Santa Marta 3, 50139 Firenze, Italy Email: paolo AT dsi DOT unifi DOT it http://www.dsi.unifi.it/neural/ From fiser at bcs.rochester.edu Sun Feb 8 15:34:50 2004 From: fiser at bcs.rochester.edu (Jozsef Fiser) Date: Sun, 8 Feb 2004 15:34:50 -0500 Subject: ECVP 2004 - Budapest, Hungary CFP Message-ID: **** ECVP 2004 **** 27th European Conference on Visual Perception Budapest (Hungary) - August 22-26, 2004 Announcement and call for papers Details of this call are now available at the direct web site of the conference: http://www.ecvp.hu or at the official web site of ECVP: http://www.ecvp.org/ You will find below a short version of the call, without the details of instructions to authors (available on the Web). Scope and topics ECVP is the largest European meeting dedicated to the theoretical and applied aspects of visual perception spanning the fields of psychophysics, experimental psychology, cognitive sciences, the neurosciences (neuroanatomy, neurophysiology, neuropsychology), neural networks, computer science, and ophthalmology. In accordance with this unique profile, the results presented at ECVP cover a wide range of themes in psychological, brain, and cognitive research related to vision, addressed by the means of electrophysiological, psychophysical, neural imagery, and computer simulation techniques. Each year the conference provides a forum for about 5-600 young and established scientists from Europe and other parts of the world to present their original contributions. All the abstracts of the conference are published in the journal Perception. More information about previous meetings is available at the official web site http://www.ecvp.org. Format The program of ECVP 2004 will consist of a plenary Perception Lecture, six invited symposia, talks, and poster presentations in two parallel oral sessions, accompanied by full-day poster sessions. The title of the symposia will be the following. * What do the extrastriate visual areas do? - Organizer : P. Lennie, NYU, USA * Functional Brain Imaging of Form and Motion Perception - Organizer : Z. Kourtzi, MPI, Germany * How visual development constraints adult vision - Organizer : R. Aslin, U. of Rochester, USA * Taking a second shot: Object perception across multiple fixations - Organizer : P. DeGraef, KUL, Belgium * Space, Time, and the Visual Control of Action - Organizer : M. Goodale, U. of Western Ontario, Canada * Natural colour constancy, retina or cortex? - Organizers : F. Cornelissen RUG, The Netherlands, and D. Foster UMIST, UK The list of confirmed speakers include: David Heeger, NYU, USA Matteo Carandini, Smith-Kettlewell, USA Jim Haxby, Princeton, USA Leslie Ungerleider NIH, USA Geoff Boynton Salk, USA Andrew Smith, U. of London, UK Mark Greenlee U. of Oldenburg, Germany Lars Muckli, MPI, Germany Oliver Braddick, UCL, UK Tony Movshon, NYU, USA Tony Norcia, Smith-Kettlewell, USA Shin Simojo, Caltech, USA Robert Gordon, NDSU, USA Rufin VanRullen, CNRS, France Filip Germeys, KUL, Belgium Martin Juttner, Aston University, UK David Millner, University of Durham, UK David Westwood, Dalhousie University, Canada Richard Andersen, Caltech, USA Yves Rossetti, INSERM Unit=E9 534, France Thomas Wachtler, U. of Marburg, Germany Maarten Kamermans, KNAW, The Netherlands Anya Hurlbert, U. of Newcastle, UK Steve Shevell, U. of Chicago, USA Karl Gegenfurtner, U. of Giessen, Germany Alex Wade, Smith-Kettlewell, USA Qasim Zaidi, SUNY, USA Location and timing The conference will be held in the very center of Budapest, at the University of Economic Sciences, on the banks of the river Danube. Budapest, is a uniquely exciting metropolis with a combination of geographic beauty, interesting architecture, lively cultural scene, famous cuisine and friendly locals making this city one of the most favorite tourist destinations in Europe. The end of August, when the conference is held, is one of the best times to visit the city for its pleasantly warm weather and slower pace. The conference is back-to-back to one of the major holidays of Hungary which is celebrated by a week-long cultural festivity with concerts and exhibitions and cumulates in spectacular fireworks on August 20th. The venue of ECVP 2004 and the prearranged accommodations are literally only steps away of the major attractions, the most sought after restaurants and the best bars of the city. Call for papers Submissions of original contributions to the 27th European Conference on Visual Perception (ECVP 2004) are now accepted on-line at http://www.ecvp.hu Prospective authors are invited to submit their contributions before March 1st 2004. There is a *NEW* submission procedure for ECVP 2004. Authors need to register before submitting an abstract, but authors of rejected abstracts get fully reimbursed. Authors must indicate their choice for oral or poster presentation at the submission. For poster presentation, the authors have to submit a short abstract. For oral presentations, in addition to the short abstract a single-page pdf file is also required (in contrast to previous meetings!). The page can contain any type of additional information to the submitted work, including graphs, pictures of stimuli, additional text, control analyses and discussion. If the author chooses so, the page can contain no extra information other than the title of the work, the authors' name and affiliation. The extra page will be used by the reviewers to qualify the work, and will not be published in Perception or publicized otherwise. The quantity or absence of the extra information will not be a determining factor in judging the quality of the submission. The abstract has to be self-contained and should make no reference to the extra information. Submissions that do not qualify for talks will be automatically considered for poster presentation. Details of the new submission procedure can be found at the ECVP 2004 web site. Deadlines Submission of contributions: March 1, 2004 Notification of acceptance: May, 2004 Early registration: April 15, 2004 Registration fees Early fee (until May 2004): Students and postdocs - EUR 120 Regular delegates - EUR 175 Late fee (after May 2004): Students and postdocs - EUR 170 Regular delegates - EUR 225 A limited amount of financial support will be available to presenting graduate students and postdocs, especially to those coming from Eastern-European countries. Details of application are available at the ECVP 2004 web site. Scientific Contact for ECVP 2004: Jozsef Fiser Department of Brain and Cognitive Sciences, University of Rochester E-mail: fiser at bcs.rochester.edu ECVP 2004 Conference Secretariat: ASSZISZTENCIA Conference Bureau Visegradi u. 29., H-1132 Budapest, Hungary Phone: +36 1 350-1854, +36 1 349-4567 Fax: +36 1 350-0929 E-mail: assziszt at chello.hu From sakas at hunter.cuny.edu Tue Feb 10 10:02:36 2004 From: sakas at hunter.cuny.edu (William Gregory Sakas) Date: Tue, 10 Feb 2004 10:02:36 -0500 Subject: Call for papers: Psycho-computational Models of Human Language Acquisition Message-ID: <1076425356.4028f28c66c38@ms1web.hunter.cuny.edu> **************************************************************************** Call for Papers COLING-2004 Workshop: Psycho-computational Models of Human Language Acquisition Geneva Switzerland 28 August 2004 http://www.colag.cs.hunter.cuny.edu/psychocomp/ Workshop Topic -------------- The workshop will be devoted to psychologically motivated computational models of language acquisition -- models that are compatible with research in psycholinguistics, developmental psychology and linguistics -- with particular emphasis on the acquisition of syntax. Invited panel: Learning Biases in Language Acquisition Models ---------------------------------------------------------------- Walter Daelemans, Antwerp and Tilburg Charles D. Yang, Yale Invited speaker --------------- Elan Dresher, Toronto Workshop Description and Motivation ----------------------------------- In recent decades there has been a great deal of successful research that applies computational learning techniques to emerging natural language technologies, along with many meetings, conferences and workshops in which to present such research. However, there have been few venues in which psycho- computational models of how humans acquire their native language(s) are the focus. Psycho-computational models of language acquisition are of particular interest in light of recent results in developmental psychology which suggest that very young infants are adept at detecting statistical patterns in an audible input stream. However, this begs the question of whether or not a psychologically plausible statistical learning strategy can be successfully exploited in a full- blown psycho-computational acquisition model. Although there has been a significant amount of presented research targeted at modeling the acquisition of word categories and phonology, research aimed at psychologically motivated modeling of syntax acquisition has just begun to emerge. The principal goal of the workshop is to bring together researchers who work within computational linguistics, formal learning theory, machine learning, artificial intelligence, linguistics, psycholinguistics and other fields, and who have created or are investigating computational models of language acquisition. In particular, it will provide a forum for establishing links and common themes between diverse paradigms. Although research which directly addresses the acquisition of syntax is strongly encouraged, related studies that inform research on the acquisition of syntax are also welcome. Papers are invited on, but not limited to, the following topics: * Acquisition models that contain a parsing component * Models that have a cross-linguistic or bilingual perspective * Models that address the question of learning bias in terms of innate linguistic knowledge versus statistical regularity in the input * Models that can acquire natural language word-order * Hybrid models that cross established paradigms * Models that directly make use of or can be used to evaluate existing linguistic or developmental theories in a computational framework (e.g. the principles & parameters framework or Optimality Theory) * Empirical models that make use of child-directed corpora * Formal models that incorporate psychologically plausible constraints * Comparative surveys, across multiple paradigms, that critique previously published studies Paper Length: Submissions should be no longer than 8 pages (A4 or the equivalent). High-quality short papers or extended abstracts of 4 to 5 pages are encouraged. Submission and format details are below. Lunch session: Word-order acquisition -------------------------------------- The topic of this session will be the acquisition of different natural language word-orders. The workshop will provide a common test-bed of abstract sentence patterns from word order divergent languages. The shared data contains the sentence patterns and cross-linguistic fully-specified parses for each sentence pattern. The patterns are available at: www.colag.cs.hunter.cuny.edu/grammar/data/allsentences.zip General information and a web interface for perusing the data can be found at: www.colag.cs.hunter.cuny.edu/grammar Due to the limited amount of time available to work with novel data, pilot studies are encouraged. The session will consist of short presentations and roundtable discussion. Submissions for this session are limited to 2 pages. Those who may be interested in submitting to this session should contact the workshop organizer before the submission deadline for further details. Dates of submissions Submission deadline: 30 March 2004 Acceptance notification: 14 May 2004 Camera-ready deadline: 10 June 2004 Workshop date: 28 August 2004 Workshop Organizer William Gregory Sakas, City University of New York (sakas at hunter.cuny.edu) Program Committee * Robert Berwick, MIT, USA * Antal van den Bosch, Tilburg University, The Netherlands * Ted Briscoe, University of Cambridge, UK * Damir Cavar, Indiana University, USA * Morten H. Christiansen, Cornell University, USA * Stephen Clark, University of Edinburgh, UK * James Cussens, University of York, UK * Walter Daelemans, University of Antwerp, Belgium and Tilburg University, The Netherlands * Jeffrey Elman, University of California, San Diego, USA * Janet Dean Fodor, City University of New York, USA * Gerard Kempen, Leiden University, The Netherlands and The Max Planck Institute, Nijmegen * Vincenzo Lombardo, University of Torino, Italy * Larry Moss, University of Indiana, USA * Miles Osborne, University of Edinburgh, UK * Dan Roth, University of Illinois at Urbana-Champaign, USA * Ivan Sag, Stanford University, USA * Jeffrey Siskind, Purdue University, USA * Mark Steedman, University of Edinburgh, UK * Menno van Zaanen, Tilburg University, The Netherlands * Charles Yang, Yale University, USA Paper Submission ---------------- Length: Submissions should be no more than 8 pages (A4 or equivalent). High- quality short papers or extended abstracts of 4 to 5 pages are encouraged. Submissions to the lunch session on word-order should be up no more that 2 pages. (If accepted, final camera ready versions may be up to 8 pages or 5 pages for the word-order submissions.) Layout: Papers must conform to COLING 2004 formatting guidelines, available at: http://www.issco.unige.ch/coling2004/coling2004downloads.html Electronic Submission: All submissions will be by email. Reviews will be blind, so be careful not to disclose authorship or affiliation. PDF submissions are preferred and will be required for the final camera-ready copy. Submissions should be sent as an attachment to: Psycho.Comp at hunter.cuny.edu. The subject line must contain the single word: Submission. Please be sure to include accurate contact information in the body of the email. Contact: Psycho.Comp at hunter.cuny.edu or sakas at hunter.cuny.edu http://www.colag.cs.hunter.cuny.edu/psychocomp/ From nestor.parga at uam.es Wed Feb 11 11:33:22 2004 From: nestor.parga at uam.es (Nestor Parga) Date: Wed, 11 Feb 2004 17:33:22 +0100 Subject: Paper available:"Role of synaptic filtering on the firing response of simple model neurons" Message-ID: <402A5952.3060500@uam.es> Dear colleagues, We would like to announce the following paper: "Role of synaptic filtering on the firing response of simple model neurons" by: Rubn Moreno-Bote and Nstor Parga (Phys. Rev. Lett, Jan 2004) A version of this work is available at: http://ket.ft.uam.es/~neurociencia/publications.html Best regards Nestor Parga ---------------------------------------------------- Title: Role of synaptic filtering on the firing response of simple model neurons Authors: Rubn Moreno-Bote and Nstor Parga Abstract: During active states of the brain neurons process their afferent currents with an effective membrane time constant much shorter than its value at rest. This fact, together with the existence of several synaptic time scales, determines to which aspects of the input the neuron responds best. Here we present a solution to the response of a leaky integrate-and-fire neuron with synaptic filters when long synaptic times are present, and predict the firing rate for all values of the synaptic time constant. We also discuss under which conditions this neuron becomes a coincidence detector. ------------------------------------------------------------------- | Nestor Parga | | | | Phone : (+34) 91-497-4542 | | Dpto. de Fisica Teorica, C-XI | Fax : (+34) 91-497-3936 | | Universidad Autonoma de Madrid | E-mail: nestor.parga at uam.es | | 28049 Madrid, SPAIN | | | | | http://ket.ft.uam.es/~neurociencia/nestor | ------------------------------------------------------------------- From tt at cs.dal.ca Wed Feb 11 16:55:37 2004 From: tt at cs.dal.ca (Thomas Trappenberg) Date: Wed, 11 Feb 2004 17:55:37 -0400 Subject: Ph.D. studentships in Canada Message-ID: <001601c3f0e9$c84cd2f0$ba43ad81@ttoffice> Ph.D. Studentship in machine learning and computational neuroscience in Canada We invite applications for Ph.D. studentships in Computer Science at Dalhousie University in Halifax, Canada (http://www.cs.dal.ca ). Several Ph.D. studentships are available in our faculty, among them two Ph.D. positions in my lab, one in data modeling and one in computational neuroscience. The research in data modeling is based on statistical and machine learning techniques with an emphasis on feature selection and classification. The research in computational neuroscience includes modeling of specific brain functions, in particular episodic memory and neural plasticity. The later research includes opportunities to work closely with members of the Dalhousie Neuroscience Institute. Please send email to tt at cs.dal.ca if you require further information. General application information is available at http://www.cs.dal.ca/~gradweb/gradSite/grad.html#Application . Regards, Thomas ------------------------------------------------- Dr. Thomas P. Trappenberg Associate Professor Faculty of Computer Science Dalhousie University 6050 University Avenue Halifax, Nova Scotia Canada B3H 1W5 Phone: (902) 494-3087 Fax: (902) 492-1517 Email: tt at cs.dal.ca From d.mareschal at bbk.ac.uk Tue Feb 10 07:37:58 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Tue, 10 Feb 2004 12:37:58 +0000 Subject: Faculty Position in London Message-ID: Dear all, The following position may be of interest to readers of this list. PLEASE DO NOT RESPOND DIRECTLY TO ME! cheers. =========================== School of Psychology Faculty of Science (Scale A/B) - Ref no: APS109 The School (RAE 5) is seeking candidates whose expertise falls in one of the following areas: cognitive psychology, developmental psychology or cognitive neuroscience. Teaching responsibilities will include research methods at undergraduate level. With a PhD, or equivalent in research and publications, in Psychology or related areas, you should have outstanding research ability or potential. The post is tenable from 1 October 2004. Salary range: =A324,325 - =A335,813 pa inc depending on qualifications and experience. Closing date: 1 March 2004. Full application details from http://www.bbk.ac.uk or email: humanresources at bbk.ac.uk or send an (A4) sae to the HR Department, Birkbeck, Malet Street, Bloomsbury, London WC1E 7HX (quoting the reference number APS109) Further Details see http://jobs.ac.uk/jobfiles/GE040.html -- ================================================== Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================== From bogus@does.not.exist.com Wed Feb 11 05:58:25 2004 From: bogus@does.not.exist.com () Date: Wed, 11 Feb 2004 11:58:25 +0100 Subject: ESANN'2004 programme ( European Symposium on Artificial Neural Networks) Message-ID: ---------------------------------------------------- | | | ESANN'2004 | | | | 11th European Symposium | | on Artificial Neural Networks | | | | Bruges (Belgium) - April 28-29-30, 2004 | | | | Preliminary program | ---------------------------------------------------- The preliminary program of the ESANN'2004 conference is now available on the Web: http://www.dice.ucl.ac.be/esann For those of you who maintain WWW pages including lists of related ANN sites: we would appreciate if you could add the above URL to your list; thank you very much! We try as much as possible to avoid multiple sendings of this message; however please apologize if you receive this e-mail twice, despite our precautions. For 12 years the ESANN conference has become a major event in the field of neural computation. ESANN is a human-size conference focusing on fundamental aspects of artificial neural networks (theory, models, algorithms, links with statistics, data analysis, biological background,...). This year, 90 scientific communications will be presented, covering most areas of the neural computation field. The programme of the conference can be found at the URL http://www.dice.ucl.ac.be/esann, together with practical information about the conference venue, registration,... Other information can be obtained by sending an e-mail to esann at dice.ucl.ac.be . ======================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - Microelectronics Laboratory 3, pl. du Levant - B-1348 Louvain-la-Neuve - Belgium tel: +32 10 47 25 51 - fax: + 32 10 47 25 98 mailto:esann at dice.ucl.ac.be * Conference secretariat d-side conference services 24 av. L. Mommaerts - B-1140 Evere - Belgium tel: + 32 2 730 06 11 - fax: + 32 2 730 06 00 mailto:esann at dice.ucl.ac.be ======================================================== From maneesh at gatsby.ucl.ac.uk Thu Feb 12 02:08:23 2004 From: maneesh at gatsby.ucl.ac.uk (Maneesh Sahani) Date: Thu, 12 Feb 2004 07:08:23 +0000 (GMT) Subject: Graduate study at the Gatsby Computational Neuroscience Unit Message-ID: <20040212070823.31D402C36@alpha.gatsby.ucl.ac.uk> Gatsby Computational Neuroscience Unit 4 year PhD Programme The Gatsby Unit is a world-class centre for theoretical neuroscience and machine learning, focusing on unsupervised learning, reinforcement learning, neural dynamics, population coding, interpretation of neural data, perceptual processing, and computational motor control. It provides a unique opportunity for a critical mass of theoreticians to interact closely with each other, and with other world-class research groups in related departments at University College London, including Anatomy, Computer Science, Functional Imaging Laboratory, Physics, Physiology, Psychology, Neurology, Ophthalmology, and Statistics. The Unit always has openings for exceptional PhD candidates. Applicants should have a strong analytical background, a keen interest in neuroscience and a relevant first degree, for example in Computer Science, Engineering, Mathematics, Neuroscience, Physics, Psychology or Statistics. The PhD programme lasts four years, including a first year of intensive instruction in techniques and research in theoretical neuroscience and machine learning. A number of competitive fully-funded studentships are available each year and the Unit also welcomes students with pre-secured funding or with other scholarship/studentship applications in progress. Applicants are encouraged to apply informally in the first instance by sending, in plain text format, a CV, a statement of research interests, and the names and addresses of three referees to admissions at gatsby.ucl.ac.uk. General enquiries should also be directed to this e-mail address. For further details of research interests please see http://www.gatsby.ucl.ac.uk/research.html Applications to begin the programme in September 2004 should be received by the 1st of March 2004. From d.mareschal at bbk.ac.uk Tue Feb 10 07:29:34 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Tue, 10 Feb 2004 12:29:34 +0000 Subject: Phd Studentships Message-ID: Readers of this list may be interested in the following Phd positions. PLEASE DO NOT RESPOND DIRECTLY TO ME. --------------------------- The School of Psychology, Birkbeck University of London has a number of Phd Studentships on offer for Phds starting in October 2004. Birkbeck College is part of the University of London and is situated in the central Bloomsbury area of London, in close proximity to University College London, The Institute of Cognitive Neuroscience, the Gatsby Computational Neurosciences Unit, the Institute of Child Health, and the Institute of Education. The School of Psychology has a very active internationally recognised research programme with particular interests in cognitive sciences, cognitive neurosciences, computational neuroscience, and cognitive and social development. However, the School welcomes applications for studentships in all areas of psychology The school of Psychology also incorporates the Centre for Brain and Cognitive Development, the Centre for Psychosocial Studies, the Institute for the Study of Children, Families & social Issues For more information about the Schools research profile and studentships available, please visit our website: http://www.psyc.bbk.ac.uk OR contact: Ms Mina Daniel Postgraduate Administrator Tel.: 020 7631 6862 E-mail: s.daniel at psychology.bbk.ac.uk -- ================================================= Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================= From sutton at cs.ualberta.ca Sun Feb 15 14:55:37 2004 From: sutton at cs.ualberta.ca (Rich Sutton) Date: Sun, 15 Feb 2004 12:55:37 -0700 Subject: ICML Workshop on Predictive Representations of World Knowledge Message-ID: Announcing an ICML workshop on PREDICTIVE REPRESENTATIONS OF WORLD KNOWLEDGE Call for participation/submissions (http://prwk.rlai.net) It has long been postulated that knowledge of the world ultimately comes down to predictions about what we will sense as a function of what we do. For example, in this view we know what is inside a box if we can predict what we would see if we were to open it. We know the three-dimensional shape of an object if we can predict its new appearance for each of various way in which we might rotate it. We might even be said to know that "Tweety is a bird" if we can make appropriate predictions about what we might see and hear (e.g., about feathers, flying, and chirps) if we were to meet Tweety. Such predictive representations of world knowledge have important potential advantages. If predictions are defined in terms of primitive sensations and actions (i.e., are grounded), then they can be directly compared with what actually does happen, enabling the knowledge to be verified or perhaps even learned without human intervention. If the predictions are deterministic, or Markov in an appropriate sense, then the knowledge can be immediately used in a variety of state-space planning methods. For these and other reasons, understanding world knowledge in predictive, sensori-motor terms has been a long-standing goal of philosophy, psychology, and artificial intelligence. So far it has remained a distant goal, but recent progress in machine learning seems to bring nearer the possibility of addressing it productively with mathematics and computational studies. In particular, we are thinking of research on: Predictive representations of state (PSRs) Observable operator models (OOMs) Temporal abstraction in reinforcement learning (options, HAMs, MAXQ) Diversity-based induction of finite-state automata Deitic, or indexical, representations Signals to symbols, symbol grounding The goal of this workshop is to bring together scientists interested in these and other topics to present, discuss, and make further progress toward understanding how knowledge of the world can be represented in predictive, sensori-motor terms. Program Committee: Rich Sutton (University of Alberta), Satinder Singh (University of Michigan), Herbert Jaeger (International University Bremen, Germany), Michael Littman (Rutgers University), Peter Stone (University of Texas at Austin), Tim Oates (University of Maryland Baltimore County), Martin Butz (University of Illinois, not confirmed) For details on submissions, due March 26, see http://prwk.rlai.net From derrick.parkhurst at jhu.edu Sun Feb 15 18:03:54 2004 From: derrick.parkhurst at jhu.edu (derrick parkhurst) Date: Sun, 15 Feb 2004 18:03:54 -0500 (EST) Subject: Graduate studentships in HCI and Psychology Message-ID: Human and Computer Vision Laboratory (HCVLab) Graduate Programs of Human Computer Interaction and Psychology Iowa State University -- Ames, Iowa The newly established HCVLab of Iowa State University is actively recruiting and seeking to support talented and motivated students wishing to undertake graduate studies leading to an M.S. or Ph.D. in Human Computer Interaction and/or Psychology. The HCVLab investigates how the human visual system receives, selects and processes information in real-world and virtual environments. Our research integrates knowledge from the fields of Psychology, Human Computer Interaction and Computer Science. We use this knowledge to innovate intelligent human computer interfaces and biologically inspired technologies. Potential areas of graduate research include: Psychology focus: The mechanisms of visual attention and human eye movements Computational modeling of visual attention and eye movements Real-world scene perception, memory and understanding Active vision and visual-motor coordination Human Computer Interaction focus: Perceptually adaptive rendering techniques in virtual reality and video compression Human computer interfaces in virtual reality, augmented reality, and mobile computing Computer vision and learning including object tracking in stereo and multi-camera vision Applicants with research experience in any of the following areas will be preferred: human vision, eye movement control, psychophysics, computational modeling, computer vision, image processing, computer graphics, virtual reality, augmented reality, and mobile computing. Applicants must hold a bachelors degree in a field such as Psychology, Computer Science, Engineering, Mathematics, Neuroscience or a closely related field. The successful candidate will be a highly motivated individual with a keen interest in addressing either basic or applied research issues related to human and/or computer vision. The successful candidate should possess strong computer skills including the ability to program in C, C++ or a similar language as well as a background in mathematics. Students may enroll in the newly established interdepartmental Human Computer Interaction Graduate Program or in the Department of Psychology Graduate Program. Terminal masters or doctoral degrees are available in Human Computer Interaction and/or Psychology. For more details about HCVLab see: http://hcvl.hci.iastate.edu/ For more details about the Graduate Program in Human Computer Interaction see: http://www.hci.iastate.edu/ For more details about the Graduate Program in Psychology see: http://www.psychology.iastate.edu/ All full-time students are eligible for competitive financial support that fully covers tuition, stipends, health insurance, and travel to research conferences. Students interested in joining the HCVLab should visit http://hcvl.hci.iastate.edu to submit a pre-application. While pre-applications will be continually reviewed, pre-applications should be submitted as soon as possible to qualify for admission in the 2004-2005 academic year. --- Derrick Parkhurst, PhD (current address) The Zanvyl Krieger Mind/Brain Institute The Johns Hopkins University Baltimore, Maryland 21218 derrick.parkhurst at jhu.edu http://cnslab.mb.jhu.edu/~derrick (address as of Summer 2004) The Department of Psychology and The Human Computer Interaction Program Iowa State University Ames, Iowa, 50011 derrick.parkhurst at hci.iastate.edu http://hcvl.hci.iastate.edu/ From nik.kasabov at aut.ac.nz Mon Feb 16 17:18:48 2004 From: nik.kasabov at aut.ac.nz (Nik Kasabov) Date: Tue, 17 Feb 2004 11:18:48 +1300 Subject: Paper on Computational neurogenetic modelling Message-ID: The following paper on Computational Neurogenetic Modelling: N.Kasabov, L.Benuskova, Computational Neurogenetics, Journal of Theoretical and Computational Nanoscience, vol.1 (1) American Scientific Publisher, 2004, in print, to appear in March, is now available as a .pdf file from: http://www.aut.ac.nz/research_showcase/research_activity_areas/kedri/downloads/pdf/KasBen_CNG-JCTNanoscience04.pdf Abstract: The aim of the paper is to introduce the scope and the problems of a new research area called Computational Neurogenetics (CNG), along with some solutions and directions for further research. CNG is concerned with the study and the development of dynamic neuronal models integrated with gene models. This area brings together knowledge from various science disciplines, such as computer and information science, neuroscience and cognitive study, genetics and molecular biology. A computational neurogenetic model is created to model a brain function or a brain disease manifestation, or to be used as a general mathematical model for solving complex scientific and engineering problems. The CNG area goes beyond modelling simple relationship between a single gene and a single neuronal function or a neuronal parameter. It is the interaction between hundreds and thousands of genes in a neuron and their relationship with the functioning of a neuronal ensemble and the brain as a whole (e.g., learning and memory, speech and vision, epilepsy, mental retardation, aging, neural stem cells, etc.). The CNG models constitute a new generation of neural network models that are closer to biological neural networks in their complex symbiosis of neuronal learning dynamics and molecular processes. Concrete models are presented as examples - evolving connectionist systems (ECOS) with evolutionary parameter optimisation and the CNG model of a class of spiking neural network ensembles (CNG-SNN). Keywords: computational neurogenetics; neural networks; gene regulatory networks; brain study; adaptive learning; computational modelling. ------------------------------------------------------------ regards Nik Kasabov Prof. Nik Kasabov, MSc, PhD FRSNZ, FNZCS, SrMIEEE Founding Director and Chief Scientist - Knowledge Engineering and Discovery Research Institute, KEDRI Chair of Knowledge Engineering, School of Computer and Information Sciences Auckland University of Technology phone: +64 9 917 9506 ; fax: +64 9 917 9501 WWW http://www.kedri.info email: nkasabov at aut.ac.nz From mvanross at inf.ed.ac.uk Tue Feb 17 05:02:47 2004 From: mvanross at inf.ed.ac.uk (Mark van Rossum) Date: Tue, 17 Feb 2004 10:02:47 +0000 (GMT) Subject: Job opening Computational Neuroscience, Edinburgh Message-ID: The School of Informatics invites applications for an appointment to a Lecturership or Readership in Neuroinformatics. You should be able to demonstrate an outstanding research record and commitment to excellence in teaching. You will be expected to play a key role in the EPSRC/MRC Doctoral Training Centre in Neuroinformatics that is hosted within the School. This unique interdisciplinary training programme encompasses computational and cognitive neuroscience, software systems and tools to support neuroscience research and neurally inspired systems including neurorobotics and neural engineering. Candidates engaged in computational modelling research are particularly encouraged to apply. You will be based in the Institute for Adaptive and Neural Computation, and will be expected to develop collaborative links and joint activities both nationally and internationally, with respect to, for example, the current UK initiatives in Systems Biology and in Cognitive Systems and the new EPSRC/MRC/BBSRC funded UK-wide Network in Neuroinformatics. We offer you the opportunity to take part in the growth of one of the leading neuroinformatics centres, as this exciting new field develops. Interaction with the University's leading researchers is encouraged, across a broad range of relevant areas within Informatics and more widely within the University, particularly with the large neuroscience community. Research within Informatics is supported by substantial funding from EPSRC, ESRC, MRC, BBRSC and the Wellcome Trust and overseas research councils, as well as through a significant portfolio of industrial collaborations and research contracts. Salary Range: ?26,270 - ?33,679 or ?35,251 - ?39,958 p.a. (under review) Application Procedure: For more information and how to apply see www.jobs.ed.ac.uk (ref 3001331) Informal enquiries about the position can be made to Professor David Willshaw (telephone: 44 131 650 4404 or e-mail: willshawinf.ed.ac.uk). From kdharris at andromeda.rutgers.edu Tue Feb 17 19:06:59 2004 From: kdharris at andromeda.rutgers.edu (Ken Harris) Date: Tue, 17 Feb 2004 19:06:59 -0500 Subject: Announcing Klusters graphical spike sorting software Message-ID: <013901c3f5b3$20ae22a0$7db5e6a5@rutgers.edu> We are pleased to announce the release of Klusters, a graphical interface for spike sorting of extracellular neuronal recordings. Klusters was written by Lynn Hazan (lynn.hazan at myrealbox.com) in the laboratory of Gyorgy Buzsaki at Rutgers University (http://osiris.rutgers.edu), in collaboration with the laboratory of Kenneth D. Harris (http://qneuro.rutgers.edu), and is released under the GNU General Public Licence. PROGRAM DESCRIPTION: -------------------- While the accuracy and speed of spike sorting is dramatically improved by automatic classification algorithms such as KlustaKwik (http://klustakwik.sourceforge.net), the results of automatic algorithms must be examined manually to ensure that the assigned clusters correspond to single neurons. Klusters has been carefully designed to make this process as fast and error-free as possible. Specific features include: *) Multiple visualization tools including waveform displays and auto- and cross-correlograms. *) "Grouping Assistant" to suggest which pairs may result from overclustering of a single neuron. *) "Recluster" feature to automatically split clusters that may comprise several distinct neurons (underclustering), using an external classification program such as KlustaKwik. *) Time trajectory display to identify possible electrode drift. *) Ergonomic user interface for maximum speed and reliability. Klusters is written in C++ for Linux under the KDE environment. The program, source code, and full documentation are available at http://klusters.sourceforge.net. ------------------------ Kenneth D. Harris, Ph.D. Assistant Professor Center for Molecular and Behavioral Neuroscience Rutgers, The State University of New Jersey 197 University Avenue Newark NJ 07102, USA phone: 973 353 1080, x3331 fax: 973 353 1272 email: kdharris at andromeda.rutgers.edu web: http://qneuro.rutgers.edu From kstanley at cs.utexas.edu Tue Feb 17 04:12:11 2004 From: kstanley at cs.utexas.edu (Kenneth Owen Stanley) Date: Tue, 17 Feb 2004 03:12:11 -0600 (CST) Subject: Paper on Evolving Increasingly Complex NN's Message-ID: <200402170912.i1H9CBVH010930@oakum.cs.utexas.edu> We are pleased to announce the publication of the following article in the Journal of Artificial Intelligence Research: Stanley, K.O. and Miikkulainen, R. (2004) "Competitive Coevolution through Evolutionary Complexification", Volume 21, pages 63-100. For quick access via your WWW browser, use this URL: http://www.jair.org/abstracts/stanley04a.html Abstract: Two major goals in machine learning are the discovery and improvement of solutions to complex problems. In this paper, we argue that complexification, i.e. the incremental elaboration of solutions through adding new structure, achieves both these goals. We demonstrate the power of complexification through the NeuroEvolution of Augmenting Topologies (NEAT) method, which evolves increasingly complex neural network architectures. NEAT is applied to an open-ended coevolutionary robot duel domain where robot controllers compete head to head. Because the robot duel domain supports a wide range of strategies, and because coevolution benefits from an escalating arms race, it serves as a suitable testbed for studying complexification. When compared to the evolution of networks with fixed structure, complexifying evolution discovers significantly more sophisticated strategies. The results suggest that in order to discover and improve complex solutions, evolution, and search in general, should be allowed to complexify as well as optimize. To see animated demos of the Robot Duel domain mentioned in the paper, please see: http://nn.cs.utexas.edu/pages/research/neatdemo.html NEAT software is available through the NEAT Page: http://nn.cs.utexas.edu/keyword?neat From golden at utdallas.edu Thu Feb 19 16:10:47 2004 From: golden at utdallas.edu (Richard Golden) Date: Thu, 19 Feb 2004 15:10:47 -0600 (CST) Subject: Invitation to Participate in Mathematical Psychology Conference Message-ID: Dear Colleague, You are invited to submit a paper for presentation at the 37th Meeting of the Society for Mathematical Psychology. The meeting will be held at the University of Michigan (July 29, 2004 through August 1, 2004). Professor Shun-ichi Amari will be giving a plenary talk at the meeting. The meeting includes a Tutorial Workshop entitled: "Differential Geometry with Applications to Measurement and Statistics" (all day July 29, 2004). The meeting will be followed by a Festschrift for Jean-Claude Falmagne on August 1-August 2! Detailed information about submitting papers for presentation at the conference and registering for the conference may be found at: http://www.cogs.indiana.edu/socmathpsych Thanks, Richard Golden Secretary-Treasurer Society for Mathematical Psychology From d.lowe at aston.ac.uk Thu Feb 19 08:18:55 2004 From: d.lowe at aston.ac.uk (D LOWE) Date: Thu, 19 Feb 2004 13:18:55 +0000 Subject: job openings at NCRG Message-ID: NCRG, Aston University, UK Lecturer in Information Engineering Over the next few months we will be looking to appoint 2 new lecturers and several postdoctoral research fellows to contribute to several funded international collaborative projects that the Group is involved in, in the broad areas of machine learning and system complexity issues that might be of interest to readers of this group. In the first instance, we are currently seeking to appoint an enthusiastic research-led individual as a full time lecturer to the Information Engineering Group within the School of Engineering and Applied Science who is also capable of contributing quality teaching to our undergraduate and graduate taught programmes which cover the mathematics provision of the School. The research areas of the Group are diverse, covering both theoretical and practical aspects of information analysis, with emphasis on the study of complex systems. Current interests include biomedical signal and pattern processing, bioinformatics, inference and graphical models, theory and algorithms of novel neural network structures, nonlinear dynamical modelling, data fusion, and the relationships between statistical physics and a variety of problems in information analysis, complexity and communications (specifically error correcting codes, steganography, cryptography and communication in distributed systems). Much of this work is in collaboration with other groups overseas and we are keen to extend this international scientific interaction. We are seeking someone who feels they can fit into the ethos of our Group and can contribute to these or related areas. Salary Range: 22,191 to 33,679, exceptionally to 37,629. Application forms and further particulars are available through our personnel web site: http://www.aston.ac.uk/hr/recruitment.htm, or by telephoning: (+44/0) 121 359 0870 (24 hour answerphone), or by email: b.a.power at aston.ac.uk, quoting reference number A04/61/99. Closing date for applications: 31 March 2004. Informal enquiries about the position, or indeed any of the upcoming positions, can be made to Professor David Lowe (d.lowe at aston.ac.uk), or Professor David Saad (d.saad at aston.ac.uk). From sfr at unipg.it Thu Feb 19 20:59:07 2004 From: sfr at unipg.it (sfr@unipg.it) Date: Fri, 20 Feb 2004 02:59:07 +0100 Subject: On blind deconvolution by neural systems Message-ID: <1077242347.403569eb9a80f@webmail.unipg.it> Dear Colleagues, I take the liberty to announce the availability of the following two new papers on blind deconvolution by neural systems: [1] "Analysis of Modified `Bussgang' Algorithms (MBA) for Channel Equalization" (by S. Fiori). To appear in the IEEE Trans. on Circuits and Systems - Part I Abstract: In our previous works, we introduced two modified `Bussgang' algorithms for blind channel equalization based on neural Bayesian iterative estimation of the source sequence. They were developed in order to reduce the computational complexity of the original `Bussgang' algorithm as well as to make it more flexible by introducing a kind of source adaptivity. However, the previous work relied on some heuristic findings, validated by series of computer-based experiments. The aim of this paper is to present a theoretical investigation of some particular aspects of the adapting equations, namely, the steady-state conditions, in order to ameliorate the performances of the modified `Bussgang' algorithms and to better explain their numerical behavior. [2] "A Fast Fixed-Point Neural Blind Deconvolution Algorithm" (by S. Fiori). To appear in the IEEE Trans. on Neural Networks Abstract: The aim of this contribution is to introduce a new blind deconvolution algorithm based on fixed-point optimization of a `Bussgang'-type cost function. The cost function relies on approximate Bayesian estimation achieved by an adaptive neuron. The main feature of the presented algorithm is fast convergence that guarantees good deconvolution performances with limited computational demand compared to algorithms of the same class. ** The draft versions of these contributions may be downloaded from the web-page http://www.unipg.it/sfr/ ("Publications" link) With best regards, Simone Fiori Faculty of Engineering - University of Perugia (Italy) Temporary visitor of Lab. for Mathematical Neuroscience (RIKEN, Japan) From gary.marcus at nyu.edu Sun Feb 22 13:24:18 2004 From: gary.marcus at nyu.edu (Gary Marcus) Date: Sun, 22 Feb 2004 13:24:18 -0500 Subject: The Birth of Mind [book announcement] Message-ID: Announcing [and with apologies for multiple postings] The Birth of the Mind How A Tiny Number of Genes Creates the Complexity of Human Thought By Gary Marcus "A joy to read." -- Publisher's Weekly "Expert and Lucid" -- Noam Chomsky "Brilliantly Original" -- Steven Pinker "[Across] such diverse disciplines as evolution, genetics, gene expression, cell biology, neurobiology, and psychology, Marcus .... makes the relevant issues understandable to the lay reader, and does an even better job of dispelling the myths that impede the way we think about genes and their role in making brains, and hence minds." -- Nature From the Jacket The Human Genome Project has blazed new trails in medical science and genetic research. We know that within hours of their birth, babies can recognize faces, connect what they hear with what they see and tell the difference between Dutch and Japanese. Our genes prepare us to observe the world; they shape the finest details of the human brain. But as far as psychology is concerned, writes award-winning cognitive scientist Gary Marcus, "it's almost as if Watson and Crick never met DNA." With The Birth of the Mind , Gary Marcus enters the nature vs. nurture debate and changes it forever. Genetics isn't destiny, but the only way to know what nature brings to the table, he argues, is to take a look at what genes actually do. Startling findings have recently revealed that the genome is much smaller than we once thought, containing no more than 30,000-40,000 genes. Since this discovery, scientists have struggled to understand how such a tiny number of genes could contain the instructions for building the human brain, arguably the most complex device in the known universe. Synthesizing up-to-the-minute research with his own original findings on child development, Marcus is the first to resolve this apparent contradiction as he chronicles exactly how genes create the infinite complexities of the human mind. Along the way, he reveals the common misconceptions people harbor about genes, and explores the stunning implications of this research for the future of genetic engineering. January 2004 (Basic Books). From ckiw at inf.ed.ac.uk Mon Feb 23 12:05:41 2004 From: ckiw at inf.ed.ac.uk (Chris Williams) Date: Mon, 23 Feb 2004 17:05:41 +0000 (GMT) Subject: Faculty Positions at the British Univeristy in Dubai Message-ID: [Note that machine learning is one of the areas highlighted. Please respond to the addresses given below.] British University in Dubai Institute of Informatics Chair and 3 Lectureships Applications are invited for the Post of Professor of Informatics and Director of the Institute, together with three lectureships in the new British University in Dubai. The British University in Dubai is a unique development in higher education in the Middle East, providing cutting edge research and education in key areas of science and technology. Each of the individual disciplinary areas of the University is strongly associated with and supported by 5/5* rated departments in leading research-intensive UK universities. The Institute of Informatics is associated with Edinburgh University and its world ranking School of Informatics. The new Professor and Director of the Institute will provide leadership in creating innovative research and teaching programmes and be involved in appointments to the lectureships. Appointment to the chair and the 3 lectureships will be made in areas of Informatics related to the first programmes to be developed by the Institute in: Natural Language Engineering, Knowledge Management & Engineering, and Machine Learning. Early research and teaching programmes will be developed in association with the University of Edinburgh's 5* rated School of Informatics. Newly appointed staff will spend part of their first year working with colleagues in Edinburgh and will be eligible for Honorary Fellowships in the University of Edinburgh. Applicants for the chair must have a strong research record. Appointment will be made on the Professorial scale of a minimum of AED 25,000 per month. Applicants for lectureships should have a PhD or equivalent, a strong and growing publication list and evidence of excellent potential in research and will be appointed on the scale AED 18,000 to AED 28,000 per month. All appointees will be eligible for a remuneration package including housing, medical, flights and set-up package. Further particulars of the posts and details of the application process can be obtained from either www.buid.ac.ae or Professor G.S.Boulton, Vice Principal, University of Edinburgh; email: Francesca.Casci at ed.ac.uk. The closing date for applications is 19 March. Interviews will be held in late March/early April. From knoblauch at neuro.informatik.uni-ulm.de Mon Feb 23 11:44:08 2004 From: knoblauch at neuro.informatik.uni-ulm.de (Andreas Knoblauch) Date: Mon, 23 Feb 2004 17:44:08 +0100 Subject: PhD thesis available: Spiking associative memory and modeling visual areas Message-ID: Dear Connectionists, My PhD thesis on "Synchronization and pattern separation in spiking associative memories and visual cortical areas" is now available on the WWW at http://www.informatik.uni-ulm.de/ni/publ/AKnoblauch/Knoblauch2003B.pdf http://www.informatik.uni-ulm.de/ni/mitarbeiter/AKnoblauch.html For those interested, I have appended the table of contents Best regards, Andreas Knoblauch. ======================================================= Keywords: associative memory, cell assemblies, spiking neural networks, spike time coding, spike synchronization, sparse distributed coding, feature binding, scene segmentation, visual cortex. table of contents: 1 Overview 1.1 Motivation and classification 1.2 Organization 2 Introduction 2.1 Neurobiology of the brain 2.2 Distributed representations, neural assemblies, and binding 2.3 Associative memory 2.4 Simulation tools and models of neurons and synapses 3 Willshaw associative memory 3.1 Classical analysis of the Willshaw model 3.2 Matrix compression yields storage capacity 1 3.3 Fault tolerance 3.4 Retrieval efficiency for sequential implementations 3.5 Retrieval efficiency for parallel implementations 3.6 A critique of the classical binomial analysis 3.7 Implications 4 Spiking associative memory 4.1 Addressing with superpositions 4.2 Pattern separation 4.3 Spike counter model 4.4 Analysis of the spike counter model 4.5 Technical implementations of the spike counter model 4.6 Biological implementations of the spike counter model 4.7 Implications 5 Model of visual cortical areas 5.1 Modeling 5.2 Interaction between two cortical areas 5.3 Relation to neurophysiological results 5.4 Summary of the mechanisms in the model 6 Binding in the brain 6.1 Is there a binding problem in the brain? 6.2 Critique of the strong Temporal Correlation Hypothesis 6.3 Sketch of a more global model of binding in cortical areas 7 Conclusions 7.1 Contributions to the theory of Willshaw associative memory 7.2 Contributions to the theory of cell assemblies 8 Zusammenfassung (in German) Appendix A Information theory A.1 Basic information theory A.2 Binary channels A.3 Optimal compression of sparse bit streams B Analysis of associative memory B.1 Information-theoretical storage capacity B.2 Auto association and hetero association B.3 Retrieval and separation quality B.4 Further analysis of Willshaw associative memory C Simulation tool Felix++ C.1 Basic architecture of Felix++ C.2 Simulation environment and components: Code examples C.3 Structure of a Felix++ simulation D Symbol reference -- --------------------------------------------------------------------------------------------------- Andreas Knoblauch Abt. Neuroinformatik Universitt Ulm James-Franck-Ring D-89069 Ulm Tel. : ++49-731-502-4255 Fax : ++49-731-502-4156 email : knoblauch at neuro.informatik.uni-ulm.de WWW : http://www.informatik.uni-ulm.de/ni/mitarbeiter/AKnoblauch.html --------------------------------------------------------------------------------------------------- From nnk at atr.co.jp Tue Feb 24 04:27:51 2004 From: nnk at atr.co.jp (Neural Networks Editorial Office) Date: Tue, 24 Feb 2004 18:27:51 +0900 Subject: Neural Networks 17(2) Message-ID: NEURAL NETWORKS 17(2) Contents - Volume 17, Number 2 - 2004 ------------------------------------------------------------------ *** NEURAL NETWORKS LETTERS *** "A computational model of spatio-temporal dynamics in depth filling-in" Shigeaki Nishina, Mitsuo Kawato "Predicting spike timings of current-injected neurons" Yasuhiro Tsuboa, Takeshi Kaneko, Shigeru Shinomoto ***** Neuroscience and Neuropsychology ***** "Configural representations in transverse patterning with a hippocampal model" Paul Rodriguez, William B. Levy "Clustering behavior in a three-layer system mimicking olivo-cerebellar dynamics" Manuel G. Velarde, Vladimir I. Nekorkin, Valeri A. Makarov, Vladimir I. Makarenko, Rodolfo R. Llina?s ***** Mathematical and Computational Analysis ***** "Information processing in a neuron ensemble with the multiplicative correlation structure" Si Wu, Shun-ichi Amari, Hiroyuki Nakahara "On-line learning through simple perceptron learning with a margin" Kazuyuki Hara, Masato Okada "Bifurcating neuron: computation and learning" Mykola Lysetskiy, Jacek M. Zuradacor "Generalized neural networks for spectral analysis: dynamics and Liapunov functions" Jose? M. Vegas, Pedro J. Zufiria "A principal components analysis self-organizing map" Ezequiel Lo?pez-Rubio, Jose? Mun??oz-Pe?rez, Jose? Antonio Go?mez-Ruiz "Regularising neural networks using flexible multivariate activation function" Mirko Solazzi, Aurelio Uncini "Neural networks for nonlinear and mixed complementarity problems and their applications" Chuangyin Dang, Yee Leung, Xing-bao Gao, Kai-zhou Chen "A canonical correlation neural network for multicollinearity and functional data" Zhenkun Gou, Colin Fyfe CURRENT EVENTS ------------------------------------------------------------------ Electronic access: www.elsevier.com/locate/neunet/. Individuals can look up instructions, aims & scope, see news, tables of contents, etc. Those who are at institutions which subscribe to Neural Networks get access to full article text as part of the institutional subscription. Sample copies can be requested for free and back issues can be ordered through the Elsevier customer support offices: nlinfo-f at elsevier.nl usinfo-f at elsevier.com or info at elsevier.co.jp ------------------------------ INNS/ENNS/JNNS Membership includes a subscription to Neural Networks: The International (INNS), European (ENNS), and Japanese (JNNS) Neural Network Societies are associations of scientists, engineers, students, and others seeking to learn about and advance the understanding of the modeling of behavioral and brain processes, and the application of neural modeling concepts to technological problems. Membership in any of the societies includes a subscription to Neural Networks, the official journal of the societies. Application forms should be sent to all the societies you want to apply to (for example, one as a member with subscription and the other one or two as a member without subscription). The JNNS does not accept credit cards or checks; to apply to the JNNS, send in the application form and wait for instructions about remitting payment. The ENNS accepts bank orders in Swedish Crowns (SEK) or credit cards. The INNS does not invoice for payment. ---------------------------------------------------------------------------- Membership Type INNS ENNS JNNS ---------------------------------------------------------------------------- membership with $80 (regular) SEK 660 Y 13,000 Neural Networks (plus Y 2,000 enrollment fee) $20 (student) SEK 460 Y11,000 (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- membership without $30 SEK 200 not available to Neural Networks non-students (subscribe through another society) Y 5,000 student (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- Name: _____________________________________ Title: _____________________________________ Address: _____________________________________ _____________________________________ _____________________________________ Phone: _____________________________________ Fax: _____________________________________ Email: _____________________________________ Payment: [ ] Check or money order enclosed, payable to INNS or ENNS OR [ ] Charge my VISA or MasterCard card number ____________________________ expiration date ________________________ INNS Membership 19 Mantua Road Mount Royal NJ 08061 USA 856 423 0162 (phone) 856 423 3420 (fax) innshq at talley.com http://www.inns.org ENNS Membership University of Skovde P.O. Box 408 531 28 Skovde Sweden 46 500 44 83 37 (phone) 46 500 44 83 99 (fax) enns at ida.his.se http://www.his.se/ida/enns JNNS Membership c/o Professor Shozo Yasui Kyushu Institute of Technology Graduate School of Life Science and Engineering 2-4 Hibikino, Wakamatsu-ku Kitakyushu 808-0196 Japan 81 93 695 6108 (phone and fax) jnns at brain.kyutech.ac.jp http://www.jnns.org/ ---------------------------------------------------------------------------- From mseeger at EECS.berkeley.EDU Tue Feb 24 14:23:21 2004 From: mseeger at EECS.berkeley.EDU (Matthias Seeger) Date: Tue, 24 Feb 2004 11:23:21 -0800 (PST) Subject: Gaussian Processes for Machine Learning Message-ID: Dear colleagues, the tutorial paper Gaussian Processes for Machine Learning will appear in the International Journal for Neural Systems (IJNS) and is available from www.cs.berkeley.edu/~mseeger/papers/bayesgp-tut.html. I hope this is interesting to some of you. Best wishes, Matthias ---- Abstract: Gaussian processes (GPs) are natural generalisations of multivariate Gaussian random variables to infinite (countably or continuous) index sets. GPs have been applied in a large number of fields to a diverse range of ends, and very many deep theoretical analyses of various properties are available. This paper gives an introduction to Gaussian processes on a fairly elementary level with special emphasis on characteristics relevant in machine learning. It draws explicit connections to branches such as spline smoothing models and support vector machines in which similar ideas have been investigated. --- Matthias Seeger Tel: 510-642-8468 485 Soda Hall, UC Berkeley Fax: 510-642-5775 Berkeley, CA 94720-1776 www.cs.berkeley.edu/~mseeger From thomas.shultz at mcgill.ca Tue Feb 24 11:55:29 2004 From: thomas.shultz at mcgill.ca (Thomas Shultz) Date: Tue, 24 Feb 2004 11:55:29 -0500 Subject: book announcement Message-ID: <000c01c3faf7$057dc320$b86ace84@psych.mcgill.ca> This book may be of interest to readers of Connectionists Computational Developmental Psychology by Thomas R. Shultz Cambridge, MA: MIT Press, 2003 $38 322 pages ISBN 0-262-19483-X Despite decades of scientific research, the core issues of child development remain too complex to be explained by traditional verbal theories. These issues include structure and transition, representation and processing, innate and experiential determinants of development, and stages of development. This book shows how computational modeling can be used to capture these complex phenomena, and lays the foundation for a new subfield of developmental psychology, computational developmental psychology. The book compares rule-based and connectionist modeling, and features constructivist models employing networks that evolve as well as learn. This allows for qualitative changes in processing mechanisms and a way to implement synaptogenesis and neurogenesis. The appendices provide additional background on the mathematical concepts used, and the book's website contains easy-to-use computational packages. Thomas R. Shultz is Professor of Psychology and Computer Science and Director of the Laboratory of Natural and Simulated Cognition at McGill University. ----------------------------------------------------------------- Thomas Shultz, Professor, Department of Psychology McGill University, 1205 Penfield Ave., Montreal, Quebec,=20 Canada H3A 1B1. Associate Member, School of Computer Science E-mail: thomas.shultz at mcgill.ca Updated 20 February 2004: http://www.psych.mcgill.ca/perpg/fac/shultz/personal/default.htm Phone: 514 398-6139 Fax: 514 398-4896 ----------------------------------------------------------------- From schunn+ at pitt.edu Wed Feb 25 11:08:36 2004 From: schunn+ at pitt.edu (Christian Schunn) Date: Wed, 25 Feb 2004 11:08:36 -0500 Subject: doctoral consortium for ICCM2004 Message-ID: NEW ANNOUNCEMENT FOR ICCM2004 The 6th International Conference on Cognitive Modeling announces a (funded) opportunity for doctoral students to attend a doctoral consortium, a small 1-day workshop just prior to the conference. The ICCM 2004 Doctoral Consortium provides an opportunity for doctoral students to explore their research interests in a multi-approach workshop, under the guidance of a panel of distinguished research faculty, consisting of David Touretzky (CMU) and Wayne Gray (RPI). The Consortium has the following objectives: * provide a setting for mutual feedback on participants' current research and guidance on future research directions * develop a supportive community of scholars and a spirit of collaborative research * contribute to the conference goals through interaction with other researchers and conference events The Consortium will be held the day before ICCM-2004, on Wednesday, July 29th, 2004. Student participants receive complimentary conference registration as well as lodging and travel reimbursement-maximum amounts will be determined at a later date. In addition to participating in the Doctoral Consortium, all Doctoral Consortium students are expected to prepare a poster for presentation at the ICCM-2004 poster session. Deadline: April 1st 2004 Review Criteria The Doctoral Consortium review committee will select participants based on their anticipated contribution to the Consortium objectives. Participants typically have settled on thesis directions but have not necessarily had their research proposals accepted by their thesis committees. The Doctoral Consortium encourages participation of students from a wide variety of modeling approaches. Students from under-represented groups or institutions, including students from institutions where modeling is not a strength, are especially encouraged to apply. Format To apply, prepare and email to iccm at pitt.edu the following: an extended abstract in PDF form and a letter of recommendation. In the email sent to iccm at pitt.edu, include the following information: your name, your department, your institution, your thesis advisor's name, and keywords that describe your research area (both modeling approach and domain area). Extended Abstract Prepare a two-page extended abstract of your thesis work in the Conference Publications Format including title, author information, abstract, keywords, thesis research summary, and references. The abstract will appear in the proceedings. Submit the abstract as a PDF file by email attachment to iccm at pitt.edu. Put your name and "Doctoral Consortium" in the email message subject field. In the email message, include your name, your department, your institution, your thesis advisor's name, and keywords that describe your research area (both modeling approach and domain area). Letter of Recommendation Have a letter of recommendation from your thesis advisor emailed to iccm at pitt.edu (by April 1st 2004). The recommendation should include formal reviews of your research and the expected timetable for completion. Posters Doctoral Consortium students are not expected to do a separate poster submission for the work that they plan to discuss at the Consortium meeting. (A submission on other work is, of course, acceptable.) Upon Acceptance Applicants will be notified by April 4th 2004. Extended abstracts of accepted submissions will be published in the ICCM 2004 Proceedings. Author Kits with instructions on how to submit materials for publication will be sent with acceptance. At the Conference Authors must be available for discussion and interaction with attendees during the scheduled poster session. Submissions * Your submission must be in English. * Submissions arriving after the deadline will not be considered. * You will receive email notification upon receipt of your submission. For more information on ICCM2004 more generally, see http://simon.lrdc.pitt.edu/~iccm/ From bressler at fau.edu Tue Feb 24 13:05:05 2004 From: bressler at fau.edu (Steven L Bressler) Date: Tue, 24 Feb 2004 13:05:05 -0500 Subject: POSTDOCTORAL POSITION AVAILABLE Message-ID: POSTDOCTORAL POSITION AVAILABLE Center for Complex Systems & Brain Sciences Florida Atlantic University A postdoctoral position is open immediately in the laboratory of Dr. Steven Bressler ( http://www.ccs.fau.edu/~bressler/ ) at the Center for Complex Systems & Brain Sciences at Florida Atlantic University ( http://www.ccs.fau.edu/ ) to participate in an ongoing research project studying the effects of attention on visuomotor processing in the cerebral cortex. The project is funded by research grants from the National Science Foundation and the National Institute of Mental Health. Desired background: -- Experience in time series analysis and multivariate statistics -- Expertise in MATLAB and/or C programming -- Experience in experimental design Please send curriculum vitae, illustrative research articles, and the names and e-mail addresses of three references to Dr. Steven Bressler at bressler at fau.edu . Steven L. Bressler, Ph.D. Professor Complex Systems & Brain Sciences Bressler Home Page From G.Brown at cs.bham.ac.uk Tue Feb 24 11:23:09 2004 From: G.Brown at cs.bham.ac.uk (Gavin Brown) Date: Tue, 24 Feb 2004 16:23:09 +0000 (GMT) Subject: PhD available: Diversity in Neural Network Ensembles In-Reply-To: References: Message-ID: Dear Connectionists, My PhD thesis on "Diversity in Neural Network Ensembles" is now available on the web at: http://www.cs.bham.ac.uk/~gxb/research/gbrown_thesis.ps.gz Abstract: --------------------------------------- We study the issue of error diversity in ensembles of neural networks. In ensembles of regression estimators, the \emph{measurement} of diversity can be formalised as the Bias-Variance-Covariance decomposition. In ensembles of classifiers, there is no neat theory in the literature to date. Our objective is to understand how to precisely define, measure, and create diverse errors for both cases. As a focal point we study one algorithm, \emph{Negative Correlation (NC) Learning} which claimed, and showed empirical evidence, to \emph{enforce} useful error diversity, creating neural network ensembles with very competitive performance on both classification and regression problems. With the lack of a solid understanding of its dynamics, we engage in a theoretical and empirical investigation. During the theoretical investigations, we find that NC succeeds due to exploiting the \emph{Ambiguity decomposition} of mean squared error. We provide a grounding for NC in a statistics context of bias, variance and covariance, including a link to a number of other algorithms that have exploited Ambiguity. The discoveries we make regarding NC are not limited to neural networks. The majority of observations we make are in fact properties of the mean squared error function. We find that NC is therefore best viewed as a \emph{framework}, rather than an algorithm itself, meaning several other learning techniques could make use of it. We further study the configurable parameter in NC, thought to be entirely problem-dependent, and find that one part of it can be analytically determined for any ensemble architecture. We proceed to define an upper bound on the remaining part of the parameter, and show considerable empirical evidence that a lower bound also exists. As the size of the ensemble increases, the upper and lower bounds converge, indicating that the optimal parameter can be determined exactly. We describe a number of experiments with different datasets and ensemble architectures, including the first comparisons to other popular ensemble methods; we find NC to be a competitive technique, worthy of further application. Finally we conclude with observations on how this investigation has impacted our understanding of diversity in general, and note several possible new directions that are suggested by this work. This includes links to evolutionary computation, Mixtures of Experts, and regularisation techniques. --------------------------------------------- Enjoy, -Gav From dalche at epigenomique.org Thu Feb 26 05:27:51 2004 From: dalche at epigenomique.org (Florence Dalche) Date: Thu, 26 Feb 2004 11:27:51 +0100 Subject: call for challenge proposals Message-ID: <403DCA27.7050902@epigenomique.org> ******************************************************************************************** 2nd CALL FOR CHALLENGE PROPOSALS of PASCAL european network of excellence (Pattern Analysis, Statistical modelling and ComputationAl Learning) http://www.pascal-network.org/challenge DEADLINE : March 4th, 2004 ******************************************************************************************** The PASCAL network of excellence invites submissions of challenge proposals for which the PASCAL joint program of challenges will provide funding and support to cover part of expenses. The aim of this challenge program is to focus attention of the research community on difficult taks that may require new algorithmic development. Both application challenge and theoretical challenges can be proposed. For application challenges,the idea is to run competitions to get no-bullshit head-to-head comparisons of established or new techniques on well chosen datasets. We expect that results will provide interesting information to a research field about comparative performance of different methods under controlled conditions. Canonical examples are the NIPS feature selection challenge (2003), the various NIST speech competitions. The proposed challenge can be a concrete task in the following fields of application of machine learning related to PASCAL: Machine vision Speech Haptics Brain computer interface Natural language processing Information Retrieval/ Textual Information Access Text-mining and understanding Multimodal integration User modelling for computer human interaction We also invite submisssions about theoretical questions and open problems relevant to pattern analysis,statistical modelling and computational learning.In this case, evaluation rules can differ and can be stated as a serie of steps to be reached. Once a proposal will be accepted, the challenge submitter will have in charge the organization and the tracking of the challenge with the help and support of the challenge program in link with the infrastructure program. Results will be presented at a challenge workshop organized at the end of the challenges. The first workshop challenge will be held in November 2004 with others evry 6 months. Web site: ======== A web site dedicated to the challenge submission process is now opened on http://www.pascal-network.org/challenge. Submission: ========== Authors of challenge proposals will be asked to answer the following questions : * Title of the proposal * Description of the challenge (including the major problem to be addressed) * Relevance with PASCAL * Expected impact on machine learning research and field of interest * Feasibility * Description/ availability of the data * Time and budget required to set up the datasets and resources * Evaluation procedure * Results already obtained (established baseline) * Estimated number of teams interested in the participation of the challenge * Indication of short term/long term challenge Key dates : ========= DEADLINE for electronic submissions : March 4th Notification to the authors : March 12th Reception of corrected electronic submissions and public notification of selected proposals: March 19th Reviewing process: ================= Each submitted challenge proposal would be reviewed by two members of the programme committee. A detailed feedback will be send back to the authors. When proposals will be accepted, there will be a short time for adjusting it to answer the comments of the reviewers from March 12th to March 19th. Preliminary programme committee: =============================== Samy Bengio (IDIAP, Martigny) Walter Daelemans (University of Antwerp) Cyril Goutte (Xerox, Grenoble) Steve Gunn (University of Southampton) Klaus-Robert Mueller (Faunhofer,Berlin) Michle Sebag (LRI, Orsay) Bill Triggs (INRIA, Grenoble) Chris Watkins (Royal Holloway university, London) Florence d'Alch-Buc (university of Paris 6) Any question regarding this call for challenge proposals may be addressed to Florence d'Alch-Buc : florence.dalche at lip6.fr with PASCAL CHALLENGE in the subject. *************************************************************************************** From frankmj at grey.colorado.edu Thu Feb 26 12:22:31 2004 From: frankmj at grey.colorado.edu (Michael J. Frank) Date: Thu, 26 Feb 2004 10:22:31 -0700 (MST) Subject: Computational Model of Cognitive Deficits in Medicated and Non-medicated Parkinsonism. Message-ID: Readers of this list may be interested in the following preprint, to appear in the Journal of Cognitive Neuroscience: http://psych.colorado.edu/~frankmj/papers/BG_DA_model.preprint.pdf Frank, M.J. (in press). Dynamic dopamine modulation in the basal ganglia: A neurocomputational account of cognitive deficits in medicated and non-medicated Parkinsonism. Journal of Cognitive Neuroscience. Abstract: Dopamine (DA) depletion in the basal ganglia (BG) of Parkinson's patients gives rise to both frontal-like and implicit learning impairments. Dopaminergic medication alleviates some cognitive deficits but impairs those that depend on intact areas of the BG, apparently due to DA ``overdose''. These findings are difficult to accommodate with verbal theories of BG/DA function, owing to complexity of system dynamics: DA dynamically modulates function in the BG, which is itself a modulatory system. This paper presents a neural network model that instantiates key biological properties and provides insight into the underlying role of DA in the BG during learning and execution of cognitive tasks. Specifically, the BG modulates the execution of ``actions'' (e.g., motor responses and working memory updating) that are being considered in different parts of frontal cortex. Phasic changes in DA, which occur during error feedback, dynamically modulate the BG threshold for facilitating/suppressing a cortical command in response to particular stimuli. Reduced dynamic range of DA explains Parkinson and DA overdose deficits with a single underlying dysfunction, despite overall differences in raw DA levels. Simulated Parkinsonism and medication effects provide a theoretical basis for behavioral data in probabilistic classification and reversal tasks. The model also provides novel testable predictions for neuropsychological and pharmacological studies, and motivates further investigation of BG/DA interactions with prefrontal cortex in working memory. From joseph at nats.informatik.uni-hamburg.de Fri Feb 27 03:23:24 2004 From: joseph at nats.informatik.uni-hamburg.de (Joseph Chen) Date: Fri, 27 Feb 2004 09:23:24 +0100 Subject: PhD Thesis on "Quantum Computation and Natural Language Processing" Message-ID: <361E265E-68FE-11D8-BC1C-003065C8C1CE@nats.informatik.uni-hamburg.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dear Connectionists, The PhD thesis: "Quantum Computation and Natural Language Processing" is available at http://www.sub.uni-hamburg.de/opus/volltexte/2002/769/pdf/ dissertation.pdf This may be of interest to some of you. Best Regards, Joseph Chen ====== Abstract ====== In this thesis, a novel approach to natural language understanding inspired by quantum mechanical principle is proposed. It is based on an analogy between the physical objects at the quantum level and human's mental states. In this way, the physical and the mental phenomena are to be understood within the same framework. It is also proposed that the apparent differences between mind and matter do not lie in the fundamental differences of their properties, but in the different manifestation of macroscopic matter and macroscopic mind owing to their different composition of pure quantum eigenstates. The apparent differences are therefore quantitative rather than qualitative. Specifically, symbols in various cognitive functions are to be treated as eigenstates with respect to a particular quantum experimental arrangement. Moreover, I claim that reasoning and inference can be treated as transformations of semiosis with symbols being the eigenstates of a particular formulation operator. The operator is the counterpart of an observable in quantum mechanics. A state of affairs (a superposition of these eigenstates) does not have well-defined physical properties until it is actually measured. Consequently the classical semantics (as classical symbols' referring to the classical physical reality) is also not well-defined and may be a misleading idea. Different from classical semantics, meaning in the quantum mechanical framework should be treated as an active measurement done on a state of affair. Moreover, the ill-definedness also manifests itself in the cognition internal to a person if we regard memory as a language-like representational system. Nevertheless, memory, treated as a specific language system, is a largely quasi-classical phenomenon in that the chemical activities in the brain are an aggregate limiting case of quantum mechanics with a very large number of quanta. The classical ``objective'' physical reality is therefore a limiting case of quantum reality as well. The general language in which common sense logic is embedded is then investigated and the apparent evasiveness and ambiguity of language can be accommodated in a quantum framework. This is done by postulating an analogous Uncertainty Principle and observing the implication of it. An important implication is the ``concept-symbol'' duality. As applications, the quantum mechanical formalism is applied to cognitive processes. For instance, non-monotonicity and counterfactual conditionals can be accommodated and assimilated in this framework. Specifically, the time-asymmetric property and the genuine unknown state of non-monotonic reasoning can be easily explained in quantum mechanics. This is also the case for the potentiality and actuality, which are crucial ideas for explaining counterfactual reasoning. Furthermore, causality can be regarded as a disguise of counterfactual reasoning. The second part of the thesis is devoted to simulations and technical applications of the aforementioned principle in natural language processing. First the preliminary experiments of common sense logic are presented. These show that the ``classicization'' of common sense logic can be implemented with very simple quantum mechanical systems. Moreover, the richness of the quantum framework goes well beyond what a classical system can offer. There can be ``fine-structures'' within seemingly simple logical arguments (XOR, for example). This is also the case for non-monotonic and counterfactual reasoning. Simple natural language tasks are also simulated based on different natural language corpora. First the syllogistic arguments embedded in natural language are simulated with a quantum system, which delivers quite remarkable results. Secondly, a monolingual syntax manipulation is implemented with a quantum system, in which the quantum mechanical approach can achieve much better performance than connectionist one. In the last experiment, a quantum mechanical architecture is trained for bilingual translation between English and German, in which there are several thorny properties in the natural language corpus, for example lexical ambiguity, separable prefixes, complicated conjugation, and non-linear translational word mappings. Nevertheless, the quantum mechanic architecture can deliver very satisfactory results. - -- Dr. Joseph C.H. Chen Computer Science Department, University of Hamburg Vogt-Koelln-Str. 30 22527 Hamburg http://nats-www.informatik.uni-hamburg.de/~joseph/ +49-40-42883-2523 (O) +49-40-42883-2515 (FAX) PGP Public Key: http://nats-www.informatik.uni-hamburg.de/~joseph/joseph.pgp.asc Printed on Planet Earth and with 100% recycled electrons. - ---------------------------------------------------------------- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (Darwin) iD8DBQFAPv6ANE0t/Ns3SDgRAtPlAJ9NMIfBPlY+aLRYnD4xyxqcp5xdUQCeO1hK NOwwv80fRi4TcJtqfdHIxQ0= =XcFl -----END PGP SIGNATURE----- From rich at ini.phys.ethz.ch Fri Feb 27 14:55:36 2004 From: rich at ini.phys.ethz.ch (Richard Hahnloser) Date: Fri, 27 Feb 2004 11:55:36 -0800 Subject: PhD Position available at the ETH in Zurich Message-ID: <403FA0B8.7030401@ini.phys.ethz.ch> Applications are invited for the position of a Doctoral Student (PhD student) to investigate novel methods of online and offline analysis of electrophysiological and acoustic data. We are a well funded research group studying the neural mechanisms of learned vocalizations in the songbird. For more information about the lab, visit http://www.ini.unizh.ch/~rich. We are seeking a motivated person with a strong background in physics and / or computer science. Research topics include but are not limited to learning algorithms for real-time song recognition and statistical methods for spike-train analysis. It will be expected of the person to closely collaborate with experimentalists in the lab. Position is available immediately for individuals with a degree in related fields. Salary is according to the guidelines of the Swiss National Science Foundation. The Institute of Neuroinformatics is a multidisciplinary institution affiliated with the Department of Physics at the Swiss Federal Institute of Technology and is located on the Irchel Campus of the University of Zurich, in the city of Zurich. For more information about the institute, visit http://www.ini.unizh.ch. Interested applicants should direct inquiry or applications with a CV and names of two references to: Dr. Richard Hahnloser Institute for Neuroinformatics, UNIZH/ETHZ Winterthurerstrasse 190 8057 Zurich Switzerland Tel: +41 1 635 3060 Email: rich at ini.phys.ethz.ch From dandre at BodyMedia.Com Mon Feb 2 14:36:19 2004 From: dandre at BodyMedia.Com (Dave Andre) Date: Mon, 2 Feb 2004 14:36:19 -0500 Subject: ICML competition and workshop on Physiological Data Modeling Message-ID: <4AA3DF1F6B40B045AC9D384A31BB7FD6126C43@mailserver.private.bodymedia.com> ICML 2004 workshop The Physiological Data Modeling Contest (PDMC) A Machine Learning Challenge July 8th, 2004 Banff, Alberta, Canada http://www.cs.utexas.edu/users/sherstov/pdmc Physiological data offers many challenges to the machine learning community including dealing with large amounts of data, sequential data, issues of sensor fusion, and a rich domain complete with noise, hidden variables, and significant effects of context. The Physiological Data Modeling Contest (PDMC) will challenge participants to exhibit machine learning algorithms for classification and regression on data taken using BodyMedia (www.bodymedia.com ) wearable body monitors. These devices collect and store continuous data from multiple sensors packaged in an unobtrusive armband. To enable this contest, BodyMedia is providing a rich dataset comprising several months of data from more than two dozen subjects. Features include raw data such as skin temperature, heat flux, galvanic skin response, and vital statistics of the user. The tasks include detecting classes of activity being engaged in by the user (e.g. walking, sleeping, watching TV, etc.) based on the values of these features. The training data is available now; the test data will be released approximately four weeks before the workshop; and the competitors will submit entries consisting of predictions for this test data shortly before ICML 2004. At the workshop, the performance of each entrant will be announced and the various techniques discussed and analyzed in detail. A small prize will be awarded to the winning competitors. Entrants will be invited to contribute papers describing their approaches to a special issue of a technical journal or magazine. The intended audience is machine learning researchers and professors currently teaching machine learning classes. Complete details are available on the contest website: http://www.cs.utexas.edu/users/sherstov/pdmc Important dates (tentative) =============== NOW: Send your expression of interest to the organizers 2 April: Notice of intent to participate due 1 June : Test data set released and announced 18 June : Entry submission deadline 8 July : Workshop/results Organizers ========== David Andre, dandre at BodyMedia.com, BodyMedia, Inc. (co-chair) Peter Stone, pstone at cs.utexas.edu, UT Austin (co-chair) Alexander Sherstov, sherstov at cs.utexas.edu, UT Austin Maxence Crossley, max at BodyMedia.com, BodyMedia, Inc. From Pierre.Bessiere at imag.fr Mon Feb 2 07:59:18 2004 From: Pierre.Bessiere at imag.fr (bessiere) Date: Mon, 02 Feb 2004 13:59:18 +0100 Subject: Bayesian Robot Programming Message-ID: We are pleased to announce the publication of the following paper in the January 2004 issue of "Autonomous Robots" a Kluewer publication: Bayesian Robot Programming ************************** Freely available at the following URL: http://www.kluweronline.com/issn/0929-5593/contents Abstract: --------- We propose a new method to program robots based on Bayesian inference and learning. It is called BRP for Bayesian Robot Programming. The capacities of this programming method are demonstrated through a succession of increasingly complex experiments. Starting from the learning of simple reactive behaviors, we present instances of behavior combination, sensor fusion, hierarchical behavior composition, situation recognition and temporal sequencing. This series of experiments comprises the steps in the incremental development of a complex robot program. The advantages and drawbacks of BRP are discussed along with these different experiments and summed up as a conclusion. These different robotics programs may be seen as an illustration of probabilistic programming applicable whenever one must deal with problems based on uncertain or incomplete knowledge. The scope of possible applications is obviously much broader than robotics. Keywords: --------- Bayesian robot programming, control of autonomous robots, computational architecture for autonomous systems, theory of autonomous systems Authors: -------- Olivier Lebeltel, Pierre Bessi?re, Julien Diard & Emmanuel Mazer ___________________________________________________________________ Dr Pierre BESSIERE ? CNRS ************************ Laboratoire GRAVIR ? Institut IMAG INRIA ? Rh?ne-Alpes 655 avenue de l?Europe 38034 St Ismier ? FRANCE Work?: +33 (0) 4.76.61.55.09 Fax?: +33 (0) 4.76.61.52.10 Mailto:Pierre.Bessiere at imag.fr WWW?: http://www-laplace.imag.fr ___________________________________________________________________ From clabiouse at ulg.ac.be Mon Feb 2 10:48:17 2004 From: clabiouse at ulg.ac.be (Christophe Labiouse) Date: Mon, 2 Feb 2004 16:48:17 +0100 Subject: LAST CALL OF PAPERS: Small Group Meeting on Social Connectionism (16-19 June 2004, Brussels, Belgium) Message-ID: <003c01c3e9a3$fa35d980$0b3ba58b@clabiouse> Small Group Meeting on Social Connectionism=20 16-19 June 2004, Brussels, Belgium [Van Overwalle (Vrije Universiteit Brussel, Belgium) and Christophe Labiouse (Universit=E9 de Li=E8ge, Belgian National Fund of Scientific Research, Belgium)] The deadline for submitting abstracts is Monday 16th February 2004. The abstract should include a list of authors, including the first author's correspondence address and email, and a summary of not more than 150 words, and should be sent to Frank Van Overwalle (Frank.VanOverwalle at vub.ac.be). =20 A number of renowned researchers in the field of connectionism will present a state-of-the-art presentation of current and novel developments in their field: Bob French (topic: dual-memory models) and Axel Cleeremans (topic: implicit learning). Moreover, renowned social researchers who worked on connectionist models in social cognition have confirmed to come to the meeting (Stephen Read and Yoshi Kashima). The meeting is limited to 20-25 participants (also PhD students or postgraduates) with an interest in social connectionism. We expect people working in the fields such as covariation, causal judgment, illusory correlation, cognitive dissonance, person perception, stereotyping, attitudes and other judgmental processes. We would also like to invite cognitive researchers or neuropsychologists working in the connectionist or associative learning domain. Topic of the meeting In the last decade, we have seen an upsurge of articles attempting to model social processes within a connectionist or computational framework. The aim of this small group meeting is to gain further insight in the possible connectionist processes that underlie social thinking by comparing and testing these different connectionist approaches. Papers are invited that propose improved models or that evaluate and discuss the merits of existing models. What are the basic properties of proposed connectionist architectures? The amount of known and novel findings in the field that it can account for? The amount of new hypothesis it generates? The breadth of empirical support? Apart from discussing the merits of specific connectionist implementations, we also want to address broader questions. What are general weaknesses of existing models? What are novel directions and phenomena that these models should attempt to cover? For instance, how are motivation, attention, and goal-directed behavior accounted for? How should we make a distinction between episodic and semantic memory? Between implicit and explicit learning and reasoning? What is the neuropsychological basis of the model's core mechanisms? Answers to these questions might improve our understanding of the connectionist underpinnings of social reasoning. Frank Van Overwalle Vrije Universiteit Brussel Department of Psychology tel: +32 2 629 25 18 Pleinlaan 2 fax: +32 2 629 24 89 B - 1050 Brussel e-mail: Frank.VanOverwalle at vub.ac.be Belgium www.vub.ac.be/PESP/VanOverwalle.html From jianfeng at sussex.ac.uk Tue Feb 3 16:23:25 2004 From: jianfeng at sussex.ac.uk (jianfeng@sussex.ac.uk) Date: Tue, 03 Feb 2004 21:23:25 +0000 Subject: postdocs In-Reply-To: References: Message-ID: <1075843405.4020114d1f7b6@webmail.sussex.ac.uk> The Babraham Institute is an independent charity whose research is applied to find new medicines and treatments for human diseases. Biological Statistician (Ref. KK/BS/ST) Starting salary range 26,000 - 33,500 p.a. A statistician with experience of multivariate analysis is required to assist in developing appropriate analytical protocols and visual representation of data from multiple cell recordings in brain sensory systems. This will also involve collaboration with a Neuroinformatics group at the University of Sussex. It is expected that the individual will subsequently apply these skills to other complex biological data and provide more general statistical advice and training for Institute scientists. The successful candidate should have a degree in biology, computing, statistics or mathematics and a higher degree in statistics or mathematics. For informal enquiries please contact Professor Keith Kendrick (keith.kendrick at bbsrc.ac.uk) Bioinformatician (Ref. WR/BIO/ST) Starting salary range 21,700 - 25,200 p.a. The bioinformatics post is at a junior level and will assist existing group members in the Bioinformatics facility with the provision of support, training and development for the research groups at the Institute. An initial focus will be for a project investigating evolutionary conservation of imprinted genes (in collaboration with the Sanger Institute). This position would suit someone wishing to pursue more formally their existing interest in bioinformatics. The successful candidate should have a degree in a Biological science. For informal enquiries please contact Dr Wolf Reik (wolf.reik at bbsrc.ac.uk) It is anticipated that funding will be available for 3 years initially for both positions. Starting salary for both posts will be according to qualifications and experience. Benefits include pension scheme. Further details and application forms available from the Personnel Office, The Babraham Institute, Babraham, Cambridge CB2 4AT. Tel. 01223 496000, e-mail: babraham.personnel at bbsrc.ac.uk. Closing date for completed applications is 27th February 2004. Please quote the appropriate reference number when applying. AN EQUAL OPPORTUNITIES EMPLOYER An Institute supported by the Biotechnology and Biological Sciences Research Council From terry at salk.edu Tue Feb 3 13:04:53 2004 From: terry at salk.edu (Terry Sejnowski) Date: Tue, 3 Feb 2004 10:04:53 -0800 (PST) Subject: NEURAL COMPUTATION 16:3 In-Reply-To: <200401080215.i082FOu12798@purkinje.salk.edu> Message-ID: <200402031804.i13I4rE81136@purkinje.salk.edu> Neural Computation - Contents - Volume 16, Number 3 - March 1, 2004 ARTICLE Geometrical Computations Explain Projection Patterns of Long-Range Horizontal Connections in Visual Cortex Ohad Ben-Shahar and Steven Zucker NOTES Mean Instantaneous Firing Frequency is Always Higher than the Firing Rate Petr Lansky, Roger Rodriguez and Laura Sacerdote Kalman Filter Control Embeded into the Reinforcement Learning Framework Istvan Szita and Andras Lorincz LETTERS Rapid Processing and Unsupervised Learning in a Model of the Cortical Macrocolumn Jorg Luecke and Christoph von der Malsburg Temporally Asymmetric Learning Supports Sequence Processing in Multi-Winner Self-Organizing Maps Reiner Schultz and James A. Reggia Neuronal Bases of Perceptual Learning Revealed by a Synaptic Balance Scheme Osamu Hoshino How the Shape of Pre-and Postsynaptic Signals Can Influence STDP: A Biophysical Model Ausra Saudargiene, Bernd Porr and Florentin Worgotter Self-Organizing Dual Coding Based on Spike-Time-Dependent Plasticity Naoki Masuda and Kazuyuki Aihara ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2004 - VOLUME 16 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $108 $54 $57.78 Individual $95 $101.65 $143 $85 $90.95 Institution $635 $679.45 $689 $572 $612.04 * includes 7% GST MIT Press Journals, 5 Cambridge Center, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu ----- From ahu at cs.stir.ac.uk Wed Feb 4 18:28:05 2004 From: ahu at cs.stir.ac.uk (Dr. Amir Hussain) Date: Wed, 4 Feb 2004 23:28:05 -0000 Subject: Final CFP: BICS'2004: Upcoming (Extended) Deadline 29 February 2004 Message-ID: <000101c3eb76$9a0f0610$1f95fc3e@DrAmir> Brain Inspired Cognitive Systems (BICS'2004): August 29 - September 1, 2004, University of Stirling, Scotland Final Call for Papers: UPCOMING (EXTENDED) PAPER SUBMISSION DEADLINE February 29, 2004 NEW: Authors of selected papers will be invited to submit an extended version of their paper for publication in a special issue of the Neurocomputing Journal, published by Elsevier Science B.V. (http://www.elsevier.nl/locate/neucom) Why This Conference, and who should Attend: The science of neural computation focuses on mathematical aspects for solving complex practical problems. It also seeks to help neurology, brain theory and cognitive psychology in the understanding of the functioning of the nervous system by means of computational models of neurons, neural nets and sub-cellular processes. BICS2004 aims to become a major point of contact for research scientists, engineers and practitioners throughout the world in the fields of cognitive and computational systems inspired by the brain and biology. Participants will share the latest research, developments and ideas in the wide arena of disciplines encompassed under the heading of BICS2004: First International ICSC Symposium on Cognitive Neuro Science (CNS 2004) (from computationally inspired models to brain-inspired computation) Chair: Prof. Igor Aleksander, Imperial College London, U.K Second International ICSC Symposium on Biologically Inspired Systems (BIS 2004) Chair: Prof. Leslie Smith, University of Stirling, U.K. Third International ICSC Symposium on Neural Computation (NC'2004) Chair: Dr. Amir Hussain, University of Stirling, U.K. http://www.icsc-naiso.org/conferences/bics2004/bics-cfp.html - Dr. Amir Hussain Dept. of Computing Science University of Stirling Stirling FK9 4LA, UK Email: ahu at cs.stir.ac.uk http://www.cs.stir.ac.uk/~ahu Tel/Fax: (+44) 01786 - 467437 / 464551 -- From bernhard.schoelkopf at tuebingen.mpg.de Thu Feb 5 03:45:28 2004 From: bernhard.schoelkopf at tuebingen.mpg.de (Bernhard Schoelkopf) Date: Thu, 5 Feb 2004 09:45:28 +0100 Subject: job openings in Tuebingen: group leaders, PhD student Message-ID: <015201c3ebc4$686458b0$ff85260a@ebro> Dear Connectionists, 1. There is a job opening for three research group leaders (C3/W2, corresponding to Associate Professor) in the Miescher Lab of the Max Planck Society, broad enough to cover the whole spectrum of the connectionists list. The Miescher Lab is on our campus, so there would be ample opportunity to interact. For details, see http://www.kyb.tuebingen.mpg.de/misc/jobs/pics/GablerMPI-Independent.pdf From ted.carnevale at yale.edu Thu Feb 5 10:02:33 2004 From: ted.carnevale at yale.edu (Ted Carnevale) Date: Thu, 05 Feb 2004 10:02:33 -0500 Subject: 2004 NEURON Simulator Meeting Message-ID: <40225B09.4040506@yale.edu> What: the first NEURON Simulator Meeting When: 9 AM Friday - Noon Sunday, May 14-16, 2004 Where: Northwestern University, Chicago, IL Registration is now open for the first NEURON Simulator Meeting. The purpose of this meeting is to bring together people who are interested in using computational modeling in neuroscience research and education--especially current users of NEURON, and others who are interested in using it--in order to: * share knowledge about technical aspects of computational modeling * keep informed about the latest advances and "best practices" in the use of NEURON * encourage participation in NEURON's development * stimulate communication and collaboration This conference will alternate between single track sessions for presentations and discussions on topics of general interest, and multiple track breakout sessions for smaller group interactions. We invite proposals for talks, symposia, tutorials, and workshops (see "Proposals Invited" below). Registration is limited to 50 individuals on a first-come, first serve basis. For more information see http://www.neuron.yale.edu/neuron/nsm2004/nsm2004.html Proposals Invited We invite participants to propose and be responsible for talks, symposia, tutorials, or workshops. Each individual whose proposal is accepted will be eligible for reimbursement of a substantial amount of travel costs to attend the meeting. See http://www.neuron.yale.edu/neuron/nsm2004/nsm2004.html for more information. --Ted From bernhard.schoelkopf at tuebingen.mpg.de Thu Feb 5 03:45:28 2004 From: bernhard.schoelkopf at tuebingen.mpg.de (Bernhard Schoelkopf) Date: Thu, 5 Feb 2004 09:45:28 +0100 Subject: job openings in Tuebingen: group leaders, PhD student Message-ID: <015201c3ebc4$686458b0$ff85260a@ebro> Dear Connectionists, 1. There is a job opening for three research group leaders (C3/W2, corresponding to Associate Professor) in the Miescher Lab of the Max Planck Society, broad enough to cover the whole spectrum of the connectionists list. The Miescher Lab is on our campus, so there would be ample opportunity to interact. For details, see http://www.kyb.tuebingen.mpg.de/misc/jobs/pics/GablerMPI-Independent.pdf 2. Our department "Empirical Inference for Machine Learning and Perception" is looking for a PhD student candidate. Applicants should be academically outstanding and highly motivated to do basic research. For more information on us, see http://www.kyb.tuebingen.mpg.de/bs. The starting date is flexible. best regards Bernhard From neil at dcs.shef.ac.uk Fri Feb 6 06:13:20 2004 From: neil at dcs.shef.ac.uk (Neil Lawrence) Date: Fri, 6 Feb 2004 11:13:20 -0000 Subject: PhD Studentship in Machine Learning and Speech Message-ID: <004e01c3eca2$3aa58910$e108a78f@windows.dcs.shef.ac.uk> Dear Connectionists, We would like to announce the following PhD studentship sponsored by the University of Sheffield. "Learning to Recognise Speech in Everyday Noisy Environments" The student will be jointly supervised by Prof. Martin Cooke and Dr Neil Lawrence in the University of Sheffield's Department of Computer Science. The closing date for applications is 5th March 2004 For more details on the project and how to apply see http://www.dcs.shef.ac.uk/~neil/studentship.html. Regards, Martin Cooke Neil Lawrence From Michael.Berthold at uni-konstanz.de Fri Feb 6 03:51:10 2004 From: Michael.Berthold at uni-konstanz.de (Michael Berthold) Date: Fri, 6 Feb 2004 09:51:10 +0100 Subject: Scholarships for PhD Program at Konstanz University Message-ID: <20040206085108.B755B1F800A@viribus.rz.uni-konstanz.de> The Department of Computer and Information Science at Konstanz University, Germany is please to announce its new PhD Graduate Program. The program is supported by the Deutsche Forschungsgemeinschaft (DFG) and offers various scholarships for graduate students and one postdoc. The topics will mainly focus around the theme of the graduate school: "Explorative Analysis and Visualization of Large Information Spaces" Please advise all interested students in your department of this great opportunity. The first deadline for applications is March 15, 2004. Detailed information about the program can be found at: http://www.inf.uni-konstanz.de/gk -- Michael R. Berthold, Professor phone: +49 (07531) 88-2202 ALTANA-Chair for Applied Computer Science fax: +49 (07531) 88-5132 (Bioinformatics and Information Mining) berthold at inf.uni-konstanz.de University of Konstanz, Germany www.inf.uni-konstanz.de/~berthold From bogus@does.not.exist.com Fri Feb 6 15:06:40 2004 From: bogus@does.not.exist.com () Date: 06 Feb 2004 15:06:40 -0500 Subject: No subject Message-ID: <1076098000.11288.218.camel@dijkstra.cs.wlu.edu> This CFP is also available in MSWord and PDF format, from: http://www.cs.wlu.edu/~levy/aaai04/AAAI_FSS04_CCCS_CFP.doc http://www.cs.wlu.edu/~levy/aaai04/AAAI_FSS04_CCCS_CFP.pdf Compositional Connectionism in Cognitive Modeling (AAAI 2004 Fall Symposium Series) October 22-24, 2004 Washington, D.C. U.S.A. This symposium will bring together connectionist and non-connectionist researchers to discuss and debate a topic of central concern in AI and cognitive science: the nature of compositionality. The open-ended productivity of the human capabilities aspired to by AI (e.g., perception, cognition, and language) is generally taken to be a consequence of compositionality; i.e., the ability to combine constituents recursively. The aim of this symposium is to expose connectionist researchers to the broadest possible range of conceptions of composition including those conceptions that pose the greatest challenge for connectionism while simultaneously alerting other AI and cognitive science researchers to the range of possibilities for connectionist implementation of composition. We therefore welcome and encourage submissions from both proponents and critics of connectionist representations, so long as the work described focuses on compositionality in an AI or in the modeling of cognition. Specifically, the symposium will seek to address the following issues: (1) What do we mean by compositional in terms of the relationship between the constituents and the composite? (2) What do we mean by the process of composition? (3) Are composition and decomposition (of a given type) completely symmetric operations? (4) What constituents (that are relevant to human agency) can be composed? (5) What are the consequences of having multiple interacting compositional systems? (6) To what extent do "structure-in-time" connectionist models, like Elman's Simple Recurrent Networks, represent compositional structure? (7) Do different connectionist models implement different aspects of compositionality, and can they be used in conjunction to address different aspects of composition? Plenary Speakers Dr. Jeffrey Elman, Dept. of Cognitive Science, University of California at San Diego Dr. Mark Steedman, School of Informatics, University of Edinburgh Submissions Standard talks at the symposium will be allocated 20 minutes for presentation, and papers will be limited to four pages. As the participants are expected to come from a wide range of disciplines, authors are asked to consider very carefully these constraints of brevity and variable background knowledge in planning their papers. Such constraints will be a primary criterion for evaluating submissions. Send submissions (in PDF, Postscript, or Microsoft Word format) to levys at wlu.edu (Simon D. Levy). Important Deadlines for Authors May 3, 2004 Submission due to organizers May 24, 2004 Notifications of acceptance sent by organizers August 31, 2004 Accepted camera-ready copy due to AAAI Registration The symposium will be limited to 40-60 participants. Those interested in attending should submit a paper (details above) or, if not intending to speak, submit a brief position paper or research abstract. Both sorts of submission (full paper or position / abstract) are due May 3, 2004. The symposium is intended for active participants, however there may be a limited number of places for interested individuals. These will be available on a first-come, first-served basis after the opening of general registration. General registration information will be available on the AAAI web site in July 2004. More detailed registration information will be available later and a separate Call for Participation will be made. Organizing Committee Simon D. Levy Computer Science Dept., Washington & Lee University Ross Gayler La Trobe University / Baycorp Advantage Value Solutions Pentti Kanerva Redwood Neuroscience Institute Chris Eliasmith Depts. of Philosophy / Systems Design Engineering, U. Waterloo Further Information For further information on the AAAI Symposia please see: http://www.aaai.org/Symposia/symposia.html http://www.aaai.org/Symposia/Fall/2004/fss-04.html -- Simon D. Levy Assistant Professor Computer Science Department Washington & Lee University Lexington, VA 24450 540-458-8419 (voice) 540-458-8479 (fax) levys at wlu.edu http://www.cs.wlu.edu/~levy From s.nolfi at istc.cnr.it Fri Feb 6 09:06:26 2004 From: s.nolfi at istc.cnr.it (Stefano Nolfi) Date: Fri, 6 Feb 2004 15:06:26 +0100 Subject: Special issue on Evolutionary Robotics Message-ID: <00e301c3ecba$68f2ca40$91419296@NOLFI> Call for Papers Special Issue on "Evolutionary Robotics: Looking forward" Connection Science journal Guest Editor : Stefano Nolfi Over the last ten years, or more, Evolutionary Robotics has attracted the interest of a large community of researchers with different research interests and backgrounds, ranging from AI and robotics, over biology and cognitive science, to the study of social behavior. Continuous progress in evolutionary robotics has led to a substantial maturation of the field and a clearer understanding of its potential and of its current limitations. The goal of this special issue is to encourage researchers working in this field not only to present their most recent research, but also to consider its relationship to other research fields, and to identify promising research directions for the future. Papers are welcome addressing (but not limited to) the following topics: Body and brain co-evolution, new materials Collective robotics and self-assembling Incremental and competitive evolution Open-ended evolutionary processes Evolution and evolvability Evolution, development and learning Representation and behavior Communication and language Cognitive capacities Specifically, the submitted papers are expected to: (a) focus on one or more related research topics, (b) explain the importance of the topic, the open problems and the different approaches discussed in the literature, (c) discuss the implications with respect to related research fields, (d) discuss the advantages and drawbacks of the evolutionary robotics approach with respect to other methodologies. Submission Instructions All manuscripts should be emailed to the guest editor (Stefano Nolfi, s.nolfi at istc.cnr.it) by May 15, 2004. Instructions for authors are available from: http://www.tandf.co.uk/journals/authors/ccosauth.asp. Manuscripts can either be full papers or shorter research notes (up to 4000 words). Important dates: Deadline for paper submission: May 15, 2004 Notification of acceptance: August 1, 2004 Final manuscript due: September 1, 2004 Publication of the special issue: December, 2004 Guest Editor: Stefano Nolfi Institute of Cognitive Sciences and Technology, CNR Viale Marx, 15 - 00137 - Rome - Italy e-mail: s.nolfi at istc.cnr.it voice: 0039-06-86090231 fax: 0039-06-824737 About the Journal Connection Science is an interdisciplinary scientific journal with a focus on the mechanisms of adaptation, cognition and intelligent behaviour in both living and artificial systems. The traditional scope of the journal has been broadened from connectionist research and neural computing to encompass work on other adaptive methods (e.g. evolutionary computing) as well as biologically inspired techniques and algorithms in applied domains. Papers should be related either to the behaviour of humans and/or other animals or to the underlying mechanisms. Papers submitted to the journal may be practical implementations, theoretical research or philosophical discussions. The submission of robotics research papers on issues raised by the interaction of agents with the environment or with other agents is particularly encouraged. ---------------------------------------------------------------------------- Stefano Nolfi Institute of Cognitive Sciences and Technologies, CNR Viale Marx, 15 - 00137 - Rome - Italy voice: 0039-06-86090231 fax: 0039-06-824737 e-mail:s.nolfi at istc.cnr.it www: http://gral.ip.rm.cnr.it/nolfi From ebaum at fastmail.fm Sun Feb 8 10:03:34 2004 From: ebaum at fastmail.fm (Eric Baum) Date: Sun, 8 Feb 2004 10:03:34 -0500 Subject: Hayek, Learning Go, Thought? Message-ID: <16422.20422.525811.937597@localhost.localdomain> My research website is back up at a new location, revised and with new material: http://www.whatisthought.com/eric.html Included are the papers on the Hayek model of evolutionary computing, including a new one applying the model to focussed web crawling, game playing papers, including new slides on A Learning Approach to Go, and DNA computing papers. If you were linked to my old NEC website, please revise your links. Also check out http://whatisthought.com , a website for the eponymous book, with ongoing posts of new research related to thought. From dummy at dsi.unifi.it Mon Feb 9 03:31:28 2004 From: dummy at dsi.unifi.it (Machine Learning Group, Univ. Firenze) Date: Mon, 09 Feb 2004 09:31:28 +0100 Subject: Postoctoral position in machine learning Message-ID: <5AEEEADE-5ADA-11D8-ADA2-000393A37AB8@dsi.unifi.it> The Neural Networks and Machine Learning group at Universita' di Firenze (Italy) is seeking one postdoctoral researcher to join an EU funded collaborative research project focused on algorithms and architectures for learning with probabilistic logic representations. The position will be funded on the three-year project APRIL II, in the 6th framework programme (Future and Emerging Technologies). APRIL II aims at developing a theoretical framework for probabilistic logic learning and demonstrate its effectiveness on significant applications, especially in the area of bioinformatics. Program partners are Freiburg University, Germany, Imperial College London, UK, Univ. of Helsinki, Finland, INRIA Rocquencourt, France, Universita' di Firenze, Italy, Aalborg University, Denmark, and Univ. of York, UK. Tight scientific collaboration is expected among all the research partners. We are especially interested in applicants with a strong background and experience in computational logic and machine learning. Experience in other related areas such as relational learning systems, probabilistic relational models, kernel methods on discrete structures, is also desiderable. Salary will be commensurate with qualifications and experience. The position is available immediately and can be renewed for up to three years. Researchers who have recently completed their Ph.D. few months are particularly encouraged to apply. Applicants are invited to send a CV and two letters of recommendation to Prof. Paolo Frasconi Dipartimento di Sistemi e Informatica Universit? di Firenze Via di Santa Marta 3, 50139 Firenze, Italy Email: paolo AT dsi DOT unifi DOT it http://www.dsi.unifi.it/neural/ From fiser at bcs.rochester.edu Sun Feb 8 15:34:50 2004 From: fiser at bcs.rochester.edu (Jozsef Fiser) Date: Sun, 8 Feb 2004 15:34:50 -0500 Subject: ECVP 2004 - Budapest, Hungary CFP Message-ID: **** ECVP 2004 **** 27th European Conference on Visual Perception Budapest (Hungary) - August 22-26, 2004 Announcement and call for papers Details of this call are now available at the direct web site of the conference: http://www.ecvp.hu or at the official web site of ECVP: http://www.ecvp.org/ You will find below a short version of the call, without the details of instructions to authors (available on the Web). Scope and topics ECVP is the largest European meeting dedicated to the theoretical and applied aspects of visual perception spanning the fields of psychophysics, experimental psychology, cognitive sciences, the neurosciences (neuroanatomy, neurophysiology, neuropsychology), neural networks, computer science, and ophthalmology. In accordance with this unique profile, the results presented at ECVP cover a wide range of themes in psychological, brain, and cognitive research related to vision, addressed by the means of electrophysiological, psychophysical, neural imagery, and computer simulation techniques. Each year the conference provides a forum for about 5-600 young and established scientists from Europe and other parts of the world to present their original contributions. All the abstracts of the conference are published in the journal Perception. More information about previous meetings is available at the official web site http://www.ecvp.org. Format The program of ECVP 2004 will consist of a plenary Perception Lecture, six invited symposia, talks, and poster presentations in two parallel oral sessions, accompanied by full-day poster sessions. The title of the symposia will be the following. * What do the extrastriate visual areas do? - Organizer : P. Lennie, NYU, USA * Functional Brain Imaging of Form and Motion Perception - Organizer : Z. Kourtzi, MPI, Germany * How visual development constraints adult vision - Organizer : R. Aslin, U. of Rochester, USA * Taking a second shot: Object perception across multiple fixations - Organizer : P. DeGraef, KUL, Belgium * Space, Time, and the Visual Control of Action - Organizer : M. Goodale, U. of Western Ontario, Canada * Natural colour constancy, retina or cortex? - Organizers : F. Cornelissen RUG, The Netherlands, and D. Foster UMIST, UK The list of confirmed speakers include: David Heeger, NYU, USA Matteo Carandini, Smith-Kettlewell, USA Jim Haxby, Princeton, USA Leslie Ungerleider NIH, USA Geoff Boynton Salk, USA Andrew Smith, U. of London, UK Mark Greenlee U. of Oldenburg, Germany Lars Muckli, MPI, Germany Oliver Braddick, UCL, UK Tony Movshon, NYU, USA Tony Norcia, Smith-Kettlewell, USA Shin Simojo, Caltech, USA Robert Gordon, NDSU, USA Rufin VanRullen, CNRS, France Filip Germeys, KUL, Belgium Martin Juttner, Aston University, UK David Millner, University of Durham, UK David Westwood, Dalhousie University, Canada Richard Andersen, Caltech, USA Yves Rossetti, INSERM Unit=E9 534, France Thomas Wachtler, U. of Marburg, Germany Maarten Kamermans, KNAW, The Netherlands Anya Hurlbert, U. of Newcastle, UK Steve Shevell, U. of Chicago, USA Karl Gegenfurtner, U. of Giessen, Germany Alex Wade, Smith-Kettlewell, USA Qasim Zaidi, SUNY, USA Location and timing The conference will be held in the very center of Budapest, at the University of Economic Sciences, on the banks of the river Danube. Budapest, is a uniquely exciting metropolis with a combination of geographic beauty, interesting architecture, lively cultural scene, famous cuisine and friendly locals making this city one of the most favorite tourist destinations in Europe. The end of August, when the conference is held, is one of the best times to visit the city for its pleasantly warm weather and slower pace. The conference is back-to-back to one of the major holidays of Hungary which is celebrated by a week-long cultural festivity with concerts and exhibitions and cumulates in spectacular fireworks on August 20th. The venue of ECVP 2004 and the prearranged accommodations are literally only steps away of the major attractions, the most sought after restaurants and the best bars of the city. Call for papers Submissions of original contributions to the 27th European Conference on Visual Perception (ECVP 2004) are now accepted on-line at http://www.ecvp.hu Prospective authors are invited to submit their contributions before March 1st 2004. There is a *NEW* submission procedure for ECVP 2004. Authors need to register before submitting an abstract, but authors of rejected abstracts get fully reimbursed. Authors must indicate their choice for oral or poster presentation at the submission. For poster presentation, the authors have to submit a short abstract. For oral presentations, in addition to the short abstract a single-page pdf file is also required (in contrast to previous meetings!). The page can contain any type of additional information to the submitted work, including graphs, pictures of stimuli, additional text, control analyses and discussion. If the author chooses so, the page can contain no extra information other than the title of the work, the authors' name and affiliation. The extra page will be used by the reviewers to qualify the work, and will not be published in Perception or publicized otherwise. The quantity or absence of the extra information will not be a determining factor in judging the quality of the submission. The abstract has to be self-contained and should make no reference to the extra information. Submissions that do not qualify for talks will be automatically considered for poster presentation. Details of the new submission procedure can be found at the ECVP 2004 web site. Deadlines Submission of contributions: March 1, 2004 Notification of acceptance: May, 2004 Early registration: April 15, 2004 Registration fees Early fee (until May 2004): Students and postdocs - EUR 120 Regular delegates - EUR 175 Late fee (after May 2004): Students and postdocs - EUR 170 Regular delegates - EUR 225 A limited amount of financial support will be available to presenting graduate students and postdocs, especially to those coming from Eastern-European countries. Details of application are available at the ECVP 2004 web site. Scientific Contact for ECVP 2004: Jozsef Fiser Department of Brain and Cognitive Sciences, University of Rochester E-mail: fiser at bcs.rochester.edu ECVP 2004 Conference Secretariat: ASSZISZTENCIA Conference Bureau Visegradi u. 29., H-1132 Budapest, Hungary Phone: +36 1 350-1854, +36 1 349-4567 Fax: +36 1 350-0929 E-mail: assziszt at chello.hu From sakas at hunter.cuny.edu Tue Feb 10 10:02:36 2004 From: sakas at hunter.cuny.edu (William Gregory Sakas) Date: Tue, 10 Feb 2004 10:02:36 -0500 Subject: Call for papers: Psycho-computational Models of Human Language Acquisition Message-ID: <1076425356.4028f28c66c38@ms1web.hunter.cuny.edu> **************************************************************************** Call for Papers COLING-2004 Workshop: Psycho-computational Models of Human Language Acquisition Geneva Switzerland 28 August 2004 http://www.colag.cs.hunter.cuny.edu/psychocomp/ Workshop Topic -------------- The workshop will be devoted to psychologically motivated computational models of language acquisition -- models that are compatible with research in psycholinguistics, developmental psychology and linguistics -- with particular emphasis on the acquisition of syntax. Invited panel: Learning Biases in Language Acquisition Models ---------------------------------------------------------------- Walter Daelemans, Antwerp and Tilburg Charles D. Yang, Yale Invited speaker --------------- Elan Dresher, Toronto Workshop Description and Motivation ----------------------------------- In recent decades there has been a great deal of successful research that applies computational learning techniques to emerging natural language technologies, along with many meetings, conferences and workshops in which to present such research. However, there have been few venues in which psycho- computational models of how humans acquire their native language(s) are the focus. Psycho-computational models of language acquisition are of particular interest in light of recent results in developmental psychology which suggest that very young infants are adept at detecting statistical patterns in an audible input stream. However, this begs the question of whether or not a psychologically plausible statistical learning strategy can be successfully exploited in a full- blown psycho-computational acquisition model. Although there has been a significant amount of presented research targeted at modeling the acquisition of word categories and phonology, research aimed at psychologically motivated modeling of syntax acquisition has just begun to emerge. The principal goal of the workshop is to bring together researchers who work within computational linguistics, formal learning theory, machine learning, artificial intelligence, linguistics, psycholinguistics and other fields, and who have created or are investigating computational models of language acquisition. In particular, it will provide a forum for establishing links and common themes between diverse paradigms. Although research which directly addresses the acquisition of syntax is strongly encouraged, related studies that inform research on the acquisition of syntax are also welcome. Papers are invited on, but not limited to, the following topics: * Acquisition models that contain a parsing component * Models that have a cross-linguistic or bilingual perspective * Models that address the question of learning bias in terms of innate linguistic knowledge versus statistical regularity in the input * Models that can acquire natural language word-order * Hybrid models that cross established paradigms * Models that directly make use of or can be used to evaluate existing linguistic or developmental theories in a computational framework (e.g. the principles & parameters framework or Optimality Theory) * Empirical models that make use of child-directed corpora * Formal models that incorporate psychologically plausible constraints * Comparative surveys, across multiple paradigms, that critique previously published studies Paper Length: Submissions should be no longer than 8 pages (A4 or the equivalent). High-quality short papers or extended abstracts of 4 to 5 pages are encouraged. Submission and format details are below. Lunch session: Word-order acquisition -------------------------------------- The topic of this session will be the acquisition of different natural language word-orders. The workshop will provide a common test-bed of abstract sentence patterns from word order divergent languages. The shared data contains the sentence patterns and cross-linguistic fully-specified parses for each sentence pattern. The patterns are available at: www.colag.cs.hunter.cuny.edu/grammar/data/allsentences.zip General information and a web interface for perusing the data can be found at: www.colag.cs.hunter.cuny.edu/grammar Due to the limited amount of time available to work with novel data, pilot studies are encouraged. The session will consist of short presentations and roundtable discussion. Submissions for this session are limited to 2 pages. Those who may be interested in submitting to this session should contact the workshop organizer before the submission deadline for further details. Dates of submissions Submission deadline: 30 March 2004 Acceptance notification: 14 May 2004 Camera-ready deadline: 10 June 2004 Workshop date: 28 August 2004 Workshop Organizer William Gregory Sakas, City University of New York (sakas at hunter.cuny.edu) Program Committee * Robert Berwick, MIT, USA * Antal van den Bosch, Tilburg University, The Netherlands * Ted Briscoe, University of Cambridge, UK * Damir Cavar, Indiana University, USA * Morten H. Christiansen, Cornell University, USA * Stephen Clark, University of Edinburgh, UK * James Cussens, University of York, UK * Walter Daelemans, University of Antwerp, Belgium and Tilburg University, The Netherlands * Jeffrey Elman, University of California, San Diego, USA * Janet Dean Fodor, City University of New York, USA * Gerard Kempen, Leiden University, The Netherlands and The Max Planck Institute, Nijmegen * Vincenzo Lombardo, University of Torino, Italy * Larry Moss, University of Indiana, USA * Miles Osborne, University of Edinburgh, UK * Dan Roth, University of Illinois at Urbana-Champaign, USA * Ivan Sag, Stanford University, USA * Jeffrey Siskind, Purdue University, USA * Mark Steedman, University of Edinburgh, UK * Menno van Zaanen, Tilburg University, The Netherlands * Charles Yang, Yale University, USA Paper Submission ---------------- Length: Submissions should be no more than 8 pages (A4 or equivalent). High- quality short papers or extended abstracts of 4 to 5 pages are encouraged. Submissions to the lunch session on word-order should be up no more that 2 pages. (If accepted, final camera ready versions may be up to 8 pages or 5 pages for the word-order submissions.) Layout: Papers must conform to COLING 2004 formatting guidelines, available at: http://www.issco.unige.ch/coling2004/coling2004downloads.html Electronic Submission: All submissions will be by email. Reviews will be blind, so be careful not to disclose authorship or affiliation. PDF submissions are preferred and will be required for the final camera-ready copy. Submissions should be sent as an attachment to: Psycho.Comp at hunter.cuny.edu. The subject line must contain the single word: Submission. Please be sure to include accurate contact information in the body of the email. Contact: Psycho.Comp at hunter.cuny.edu or sakas at hunter.cuny.edu http://www.colag.cs.hunter.cuny.edu/psychocomp/ From nestor.parga at uam.es Wed Feb 11 11:33:22 2004 From: nestor.parga at uam.es (Nestor Parga) Date: Wed, 11 Feb 2004 17:33:22 +0100 Subject: Paper available:"Role of synaptic filtering on the firing response of simple model neurons" Message-ID: <402A5952.3060500@uam.es> Dear colleagues, We would like to announce the following paper: "Role of synaptic filtering on the firing response of simple model neurons" by: Rubn Moreno-Bote and Nstor Parga (Phys. Rev. Lett, Jan 2004) A version of this work is available at: http://ket.ft.uam.es/~neurociencia/publications.html Best regards Nestor Parga ---------------------------------------------------- Title: Role of synaptic filtering on the firing response of simple model neurons Authors: Rubn Moreno-Bote and Nstor Parga Abstract: During active states of the brain neurons process their afferent currents with an effective membrane time constant much shorter than its value at rest. This fact, together with the existence of several synaptic time scales, determines to which aspects of the input the neuron responds best. Here we present a solution to the response of a leaky integrate-and-fire neuron with synaptic filters when long synaptic times are present, and predict the firing rate for all values of the synaptic time constant. We also discuss under which conditions this neuron becomes a coincidence detector. ------------------------------------------------------------------- | Nestor Parga | | | | Phone : (+34) 91-497-4542 | | Dpto. de Fisica Teorica, C-XI | Fax : (+34) 91-497-3936 | | Universidad Autonoma de Madrid | E-mail: nestor.parga at uam.es | | 28049 Madrid, SPAIN | | | | | http://ket.ft.uam.es/~neurociencia/nestor | ------------------------------------------------------------------- From tt at cs.dal.ca Wed Feb 11 16:55:37 2004 From: tt at cs.dal.ca (Thomas Trappenberg) Date: Wed, 11 Feb 2004 17:55:37 -0400 Subject: Ph.D. studentships in Canada Message-ID: <001601c3f0e9$c84cd2f0$ba43ad81@ttoffice> Ph.D. Studentship in machine learning and computational neuroscience in Canada We invite applications for Ph.D. studentships in Computer Science at Dalhousie University in Halifax, Canada (http://www.cs.dal.ca ). Several Ph.D. studentships are available in our faculty, among them two Ph.D. positions in my lab, one in data modeling and one in computational neuroscience. The research in data modeling is based on statistical and machine learning techniques with an emphasis on feature selection and classification. The research in computational neuroscience includes modeling of specific brain functions, in particular episodic memory and neural plasticity. The later research includes opportunities to work closely with members of the Dalhousie Neuroscience Institute. Please send email to tt at cs.dal.ca if you require further information. General application information is available at http://www.cs.dal.ca/~gradweb/gradSite/grad.html#Application . Regards, Thomas ------------------------------------------------- Dr. Thomas P. Trappenberg Associate Professor Faculty of Computer Science Dalhousie University 6050 University Avenue Halifax, Nova Scotia Canada B3H 1W5 Phone: (902) 494-3087 Fax: (902) 492-1517 Email: tt at cs.dal.ca From d.mareschal at bbk.ac.uk Tue Feb 10 07:37:58 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Tue, 10 Feb 2004 12:37:58 +0000 Subject: Faculty Position in London Message-ID: Dear all, The following position may be of interest to readers of this list. PLEASE DO NOT RESPOND DIRECTLY TO ME! cheers. =========================== School of Psychology Faculty of Science (Scale A/B) - Ref no: APS109 The School (RAE 5) is seeking candidates whose expertise falls in one of the following areas: cognitive psychology, developmental psychology or cognitive neuroscience. Teaching responsibilities will include research methods at undergraduate level. With a PhD, or equivalent in research and publications, in Psychology or related areas, you should have outstanding research ability or potential. The post is tenable from 1 October 2004. Salary range: =A324,325 - =A335,813 pa inc depending on qualifications and experience. Closing date: 1 March 2004. Full application details from http://www.bbk.ac.uk or email: humanresources at bbk.ac.uk or send an (A4) sae to the HR Department, Birkbeck, Malet Street, Bloomsbury, London WC1E 7HX (quoting the reference number APS109) Further Details see http://jobs.ac.uk/jobfiles/GE040.html -- ================================================== Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================== From bogus@does.not.exist.com Wed Feb 11 05:58:25 2004 From: bogus@does.not.exist.com () Date: Wed, 11 Feb 2004 11:58:25 +0100 Subject: ESANN'2004 programme ( European Symposium on Artificial Neural Networks) Message-ID: ---------------------------------------------------- | | | ESANN'2004 | | | | 11th European Symposium | | on Artificial Neural Networks | | | | Bruges (Belgium) - April 28-29-30, 2004 | | | | Preliminary program | ---------------------------------------------------- The preliminary program of the ESANN'2004 conference is now available on the Web: http://www.dice.ucl.ac.be/esann For those of you who maintain WWW pages including lists of related ANN sites: we would appreciate if you could add the above URL to your list; thank you very much! We try as much as possible to avoid multiple sendings of this message; however please apologize if you receive this e-mail twice, despite our precautions. For 12 years the ESANN conference has become a major event in the field of neural computation. ESANN is a human-size conference focusing on fundamental aspects of artificial neural networks (theory, models, algorithms, links with statistics, data analysis, biological background,...). This year, 90 scientific communications will be presented, covering most areas of the neural computation field. The programme of the conference can be found at the URL http://www.dice.ucl.ac.be/esann, together with practical information about the conference venue, registration,... Other information can be obtained by sending an e-mail to esann at dice.ucl.ac.be . ======================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - Microelectronics Laboratory 3, pl. du Levant - B-1348 Louvain-la-Neuve - Belgium tel: +32 10 47 25 51 - fax: + 32 10 47 25 98 mailto:esann at dice.ucl.ac.be * Conference secretariat d-side conference services 24 av. L. Mommaerts - B-1140 Evere - Belgium tel: + 32 2 730 06 11 - fax: + 32 2 730 06 00 mailto:esann at dice.ucl.ac.be ======================================================== From maneesh at gatsby.ucl.ac.uk Thu Feb 12 02:08:23 2004 From: maneesh at gatsby.ucl.ac.uk (Maneesh Sahani) Date: Thu, 12 Feb 2004 07:08:23 +0000 (GMT) Subject: Graduate study at the Gatsby Computational Neuroscience Unit Message-ID: <20040212070823.31D402C36@alpha.gatsby.ucl.ac.uk> Gatsby Computational Neuroscience Unit 4 year PhD Programme The Gatsby Unit is a world-class centre for theoretical neuroscience and machine learning, focusing on unsupervised learning, reinforcement learning, neural dynamics, population coding, interpretation of neural data, perceptual processing, and computational motor control. It provides a unique opportunity for a critical mass of theoreticians to interact closely with each other, and with other world-class research groups in related departments at University College London, including Anatomy, Computer Science, Functional Imaging Laboratory, Physics, Physiology, Psychology, Neurology, Ophthalmology, and Statistics. The Unit always has openings for exceptional PhD candidates. Applicants should have a strong analytical background, a keen interest in neuroscience and a relevant first degree, for example in Computer Science, Engineering, Mathematics, Neuroscience, Physics, Psychology or Statistics. The PhD programme lasts four years, including a first year of intensive instruction in techniques and research in theoretical neuroscience and machine learning. A number of competitive fully-funded studentships are available each year and the Unit also welcomes students with pre-secured funding or with other scholarship/studentship applications in progress. Applicants are encouraged to apply informally in the first instance by sending, in plain text format, a CV, a statement of research interests, and the names and addresses of three referees to admissions at gatsby.ucl.ac.uk. General enquiries should also be directed to this e-mail address. For further details of research interests please see http://www.gatsby.ucl.ac.uk/research.html Applications to begin the programme in September 2004 should be received by the 1st of March 2004. From d.mareschal at bbk.ac.uk Tue Feb 10 07:29:34 2004 From: d.mareschal at bbk.ac.uk (Denis Mareschal) Date: Tue, 10 Feb 2004 12:29:34 +0000 Subject: Phd Studentships Message-ID: Readers of this list may be interested in the following Phd positions. PLEASE DO NOT RESPOND DIRECTLY TO ME. --------------------------- The School of Psychology, Birkbeck University of London has a number of Phd Studentships on offer for Phds starting in October 2004. Birkbeck College is part of the University of London and is situated in the central Bloomsbury area of London, in close proximity to University College London, The Institute of Cognitive Neuroscience, the Gatsby Computational Neurosciences Unit, the Institute of Child Health, and the Institute of Education. The School of Psychology has a very active internationally recognised research programme with particular interests in cognitive sciences, cognitive neurosciences, computational neuroscience, and cognitive and social development. However, the School welcomes applications for studentships in all areas of psychology The school of Psychology also incorporates the Centre for Brain and Cognitive Development, the Centre for Psychosocial Studies, the Institute for the Study of Children, Families & social Issues For more information about the Schools research profile and studentships available, please visit our website: http://www.psyc.bbk.ac.uk OR contact: Ms Mina Daniel Postgraduate Administrator Tel.: 020 7631 6862 E-mail: s.daniel at psychology.bbk.ac.uk -- ================================================= Dr. Denis Mareschal Centre for Brain and Cognitive Development School of Psychology Birkbeck College University of London Malet St., London WC1E 7HX, UK tel +44 (0)20 7631-6582/6226 reception: 6207 fax +44 (0)20 7631-6312 http://www.psyc.bbk.ac.uk/people/academic/mareschal_d/ ================================================= From sutton at cs.ualberta.ca Sun Feb 15 14:55:37 2004 From: sutton at cs.ualberta.ca (Rich Sutton) Date: Sun, 15 Feb 2004 12:55:37 -0700 Subject: ICML Workshop on Predictive Representations of World Knowledge Message-ID: Announcing an ICML workshop on PREDICTIVE REPRESENTATIONS OF WORLD KNOWLEDGE Call for participation/submissions (http://prwk.rlai.net) It has long been postulated that knowledge of the world ultimately comes down to predictions about what we will sense as a function of what we do. For example, in this view we know what is inside a box if we can predict what we would see if we were to open it. We know the three-dimensional shape of an object if we can predict its new appearance for each of various way in which we might rotate it. We might even be said to know that "Tweety is a bird" if we can make appropriate predictions about what we might see and hear (e.g., about feathers, flying, and chirps) if we were to meet Tweety. Such predictive representations of world knowledge have important potential advantages. If predictions are defined in terms of primitive sensations and actions (i.e., are grounded), then they can be directly compared with what actually does happen, enabling the knowledge to be verified or perhaps even learned without human intervention. If the predictions are deterministic, or Markov in an appropriate sense, then the knowledge can be immediately used in a variety of state-space planning methods. For these and other reasons, understanding world knowledge in predictive, sensori-motor terms has been a long-standing goal of philosophy, psychology, and artificial intelligence. So far it has remained a distant goal, but recent progress in machine learning seems to bring nearer the possibility of addressing it productively with mathematics and computational studies. In particular, we are thinking of research on: Predictive representations of state (PSRs) Observable operator models (OOMs) Temporal abstraction in reinforcement learning (options, HAMs, MAXQ) Diversity-based induction of finite-state automata Deitic, or indexical, representations Signals to symbols, symbol grounding The goal of this workshop is to bring together scientists interested in these and other topics to present, discuss, and make further progress toward understanding how knowledge of the world can be represented in predictive, sensori-motor terms. Program Committee: Rich Sutton (University of Alberta), Satinder Singh (University of Michigan), Herbert Jaeger (International University Bremen, Germany), Michael Littman (Rutgers University), Peter Stone (University of Texas at Austin), Tim Oates (University of Maryland Baltimore County), Martin Butz (University of Illinois, not confirmed) For details on submissions, due March 26, see http://prwk.rlai.net From derrick.parkhurst at jhu.edu Sun Feb 15 18:03:54 2004 From: derrick.parkhurst at jhu.edu (derrick parkhurst) Date: Sun, 15 Feb 2004 18:03:54 -0500 (EST) Subject: Graduate studentships in HCI and Psychology Message-ID: Human and Computer Vision Laboratory (HCVLab) Graduate Programs of Human Computer Interaction and Psychology Iowa State University -- Ames, Iowa The newly established HCVLab of Iowa State University is actively recruiting and seeking to support talented and motivated students wishing to undertake graduate studies leading to an M.S. or Ph.D. in Human Computer Interaction and/or Psychology. The HCVLab investigates how the human visual system receives, selects and processes information in real-world and virtual environments. Our research integrates knowledge from the fields of Psychology, Human Computer Interaction and Computer Science. We use this knowledge to innovate intelligent human computer interfaces and biologically inspired technologies. Potential areas of graduate research include: Psychology focus: The mechanisms of visual attention and human eye movements Computational modeling of visual attention and eye movements Real-world scene perception, memory and understanding Active vision and visual-motor coordination Human Computer Interaction focus: Perceptually adaptive rendering techniques in virtual reality and video compression Human computer interfaces in virtual reality, augmented reality, and mobile computing Computer vision and learning including object tracking in stereo and multi-camera vision Applicants with research experience in any of the following areas will be preferred: human vision, eye movement control, psychophysics, computational modeling, computer vision, image processing, computer graphics, virtual reality, augmented reality, and mobile computing. Applicants must hold a bachelors degree in a field such as Psychology, Computer Science, Engineering, Mathematics, Neuroscience or a closely related field. The successful candidate will be a highly motivated individual with a keen interest in addressing either basic or applied research issues related to human and/or computer vision. The successful candidate should possess strong computer skills including the ability to program in C, C++ or a similar language as well as a background in mathematics. Students may enroll in the newly established interdepartmental Human Computer Interaction Graduate Program or in the Department of Psychology Graduate Program. Terminal masters or doctoral degrees are available in Human Computer Interaction and/or Psychology. For more details about HCVLab see: http://hcvl.hci.iastate.edu/ For more details about the Graduate Program in Human Computer Interaction see: http://www.hci.iastate.edu/ For more details about the Graduate Program in Psychology see: http://www.psychology.iastate.edu/ All full-time students are eligible for competitive financial support that fully covers tuition, stipends, health insurance, and travel to research conferences. Students interested in joining the HCVLab should visit http://hcvl.hci.iastate.edu to submit a pre-application. While pre-applications will be continually reviewed, pre-applications should be submitted as soon as possible to qualify for admission in the 2004-2005 academic year. --- Derrick Parkhurst, PhD (current address) The Zanvyl Krieger Mind/Brain Institute The Johns Hopkins University Baltimore, Maryland 21218 derrick.parkhurst at jhu.edu http://cnslab.mb.jhu.edu/~derrick (address as of Summer 2004) The Department of Psychology and The Human Computer Interaction Program Iowa State University Ames, Iowa, 50011 derrick.parkhurst at hci.iastate.edu http://hcvl.hci.iastate.edu/ From nik.kasabov at aut.ac.nz Mon Feb 16 17:18:48 2004 From: nik.kasabov at aut.ac.nz (Nik Kasabov) Date: Tue, 17 Feb 2004 11:18:48 +1300 Subject: Paper on Computational neurogenetic modelling Message-ID: The following paper on Computational Neurogenetic Modelling: N.Kasabov, L.Benuskova, Computational Neurogenetics, Journal of Theoretical and Computational Nanoscience, vol.1 (1) American Scientific Publisher, 2004, in print, to appear in March, is now available as a .pdf file from: http://www.aut.ac.nz/research_showcase/research_activity_areas/kedri/downloads/pdf/KasBen_CNG-JCTNanoscience04.pdf Abstract: The aim of the paper is to introduce the scope and the problems of a new research area called Computational Neurogenetics (CNG), along with some solutions and directions for further research. CNG is concerned with the study and the development of dynamic neuronal models integrated with gene models. This area brings together knowledge from various science disciplines, such as computer and information science, neuroscience and cognitive study, genetics and molecular biology. A computational neurogenetic model is created to model a brain function or a brain disease manifestation, or to be used as a general mathematical model for solving complex scientific and engineering problems. The CNG area goes beyond modelling simple relationship between a single gene and a single neuronal function or a neuronal parameter. It is the interaction between hundreds and thousands of genes in a neuron and their relationship with the functioning of a neuronal ensemble and the brain as a whole (e.g., learning and memory, speech and vision, epilepsy, mental retardation, aging, neural stem cells, etc.). The CNG models constitute a new generation of neural network models that are closer to biological neural networks in their complex symbiosis of neuronal learning dynamics and molecular processes. Concrete models are presented as examples - evolving connectionist systems (ECOS) with evolutionary parameter optimisation and the CNG model of a class of spiking neural network ensembles (CNG-SNN). Keywords: computational neurogenetics; neural networks; gene regulatory networks; brain study; adaptive learning; computational modelling. ------------------------------------------------------------ regards Nik Kasabov Prof. Nik Kasabov, MSc, PhD FRSNZ, FNZCS, SrMIEEE Founding Director and Chief Scientist - Knowledge Engineering and Discovery Research Institute, KEDRI Chair of Knowledge Engineering, School of Computer and Information Sciences Auckland University of Technology phone: +64 9 917 9506 ; fax: +64 9 917 9501 WWW http://www.kedri.info email: nkasabov at aut.ac.nz From mvanross at inf.ed.ac.uk Tue Feb 17 05:02:47 2004 From: mvanross at inf.ed.ac.uk (Mark van Rossum) Date: Tue, 17 Feb 2004 10:02:47 +0000 (GMT) Subject: Job opening Computational Neuroscience, Edinburgh Message-ID: The School of Informatics invites applications for an appointment to a Lecturership or Readership in Neuroinformatics. You should be able to demonstrate an outstanding research record and commitment to excellence in teaching. You will be expected to play a key role in the EPSRC/MRC Doctoral Training Centre in Neuroinformatics that is hosted within the School. This unique interdisciplinary training programme encompasses computational and cognitive neuroscience, software systems and tools to support neuroscience research and neurally inspired systems including neurorobotics and neural engineering. Candidates engaged in computational modelling research are particularly encouraged to apply. You will be based in the Institute for Adaptive and Neural Computation, and will be expected to develop collaborative links and joint activities both nationally and internationally, with respect to, for example, the current UK initiatives in Systems Biology and in Cognitive Systems and the new EPSRC/MRC/BBSRC funded UK-wide Network in Neuroinformatics. We offer you the opportunity to take part in the growth of one of the leading neuroinformatics centres, as this exciting new field develops. Interaction with the University's leading researchers is encouraged, across a broad range of relevant areas within Informatics and more widely within the University, particularly with the large neuroscience community. Research within Informatics is supported by substantial funding from EPSRC, ESRC, MRC, BBRSC and the Wellcome Trust and overseas research councils, as well as through a significant portfolio of industrial collaborations and research contracts. Salary Range: ?26,270 - ?33,679 or ?35,251 - ?39,958 p.a. (under review) Application Procedure: For more information and how to apply see www.jobs.ed.ac.uk (ref 3001331) Informal enquiries about the position can be made to Professor David Willshaw (telephone: 44 131 650 4404 or e-mail: willshawinf.ed.ac.uk). From kdharris at andromeda.rutgers.edu Tue Feb 17 19:06:59 2004 From: kdharris at andromeda.rutgers.edu (Ken Harris) Date: Tue, 17 Feb 2004 19:06:59 -0500 Subject: Announcing Klusters graphical spike sorting software Message-ID: <013901c3f5b3$20ae22a0$7db5e6a5@rutgers.edu> We are pleased to announce the release of Klusters, a graphical interface for spike sorting of extracellular neuronal recordings. Klusters was written by Lynn Hazan (lynn.hazan at myrealbox.com) in the laboratory of Gyorgy Buzsaki at Rutgers University (http://osiris.rutgers.edu), in collaboration with the laboratory of Kenneth D. Harris (http://qneuro.rutgers.edu), and is released under the GNU General Public Licence. PROGRAM DESCRIPTION: -------------------- While the accuracy and speed of spike sorting is dramatically improved by automatic classification algorithms such as KlustaKwik (http://klustakwik.sourceforge.net), the results of automatic algorithms must be examined manually to ensure that the assigned clusters correspond to single neurons. Klusters has been carefully designed to make this process as fast and error-free as possible. Specific features include: *) Multiple visualization tools including waveform displays and auto- and cross-correlograms. *) "Grouping Assistant" to suggest which pairs may result from overclustering of a single neuron. *) "Recluster" feature to automatically split clusters that may comprise several distinct neurons (underclustering), using an external classification program such as KlustaKwik. *) Time trajectory display to identify possible electrode drift. *) Ergonomic user interface for maximum speed and reliability. Klusters is written in C++ for Linux under the KDE environment. The program, source code, and full documentation are available at http://klusters.sourceforge.net. ------------------------ Kenneth D. Harris, Ph.D. Assistant Professor Center for Molecular and Behavioral Neuroscience Rutgers, The State University of New Jersey 197 University Avenue Newark NJ 07102, USA phone: 973 353 1080, x3331 fax: 973 353 1272 email: kdharris at andromeda.rutgers.edu web: http://qneuro.rutgers.edu From kstanley at cs.utexas.edu Tue Feb 17 04:12:11 2004 From: kstanley at cs.utexas.edu (Kenneth Owen Stanley) Date: Tue, 17 Feb 2004 03:12:11 -0600 (CST) Subject: Paper on Evolving Increasingly Complex NN's Message-ID: <200402170912.i1H9CBVH010930@oakum.cs.utexas.edu> We are pleased to announce the publication of the following article in the Journal of Artificial Intelligence Research: Stanley, K.O. and Miikkulainen, R. (2004) "Competitive Coevolution through Evolutionary Complexification", Volume 21, pages 63-100. For quick access via your WWW browser, use this URL: http://www.jair.org/abstracts/stanley04a.html Abstract: Two major goals in machine learning are the discovery and improvement of solutions to complex problems. In this paper, we argue that complexification, i.e. the incremental elaboration of solutions through adding new structure, achieves both these goals. We demonstrate the power of complexification through the NeuroEvolution of Augmenting Topologies (NEAT) method, which evolves increasingly complex neural network architectures. NEAT is applied to an open-ended coevolutionary robot duel domain where robot controllers compete head to head. Because the robot duel domain supports a wide range of strategies, and because coevolution benefits from an escalating arms race, it serves as a suitable testbed for studying complexification. When compared to the evolution of networks with fixed structure, complexifying evolution discovers significantly more sophisticated strategies. The results suggest that in order to discover and improve complex solutions, evolution, and search in general, should be allowed to complexify as well as optimize. To see animated demos of the Robot Duel domain mentioned in the paper, please see: http://nn.cs.utexas.edu/pages/research/neatdemo.html NEAT software is available through the NEAT Page: http://nn.cs.utexas.edu/keyword?neat From golden at utdallas.edu Thu Feb 19 16:10:47 2004 From: golden at utdallas.edu (Richard Golden) Date: Thu, 19 Feb 2004 15:10:47 -0600 (CST) Subject: Invitation to Participate in Mathematical Psychology Conference Message-ID: Dear Colleague, You are invited to submit a paper for presentation at the 37th Meeting of the Society for Mathematical Psychology. The meeting will be held at the University of Michigan (July 29, 2004 through August 1, 2004). Professor Shun-ichi Amari will be giving a plenary talk at the meeting. The meeting includes a Tutorial Workshop entitled: "Differential Geometry with Applications to Measurement and Statistics" (all day July 29, 2004). The meeting will be followed by a Festschrift for Jean-Claude Falmagne on August 1-August 2! Detailed information about submitting papers for presentation at the conference and registering for the conference may be found at: http://www.cogs.indiana.edu/socmathpsych Thanks, Richard Golden Secretary-Treasurer Society for Mathematical Psychology From d.lowe at aston.ac.uk Thu Feb 19 08:18:55 2004 From: d.lowe at aston.ac.uk (D LOWE) Date: Thu, 19 Feb 2004 13:18:55 +0000 Subject: job openings at NCRG Message-ID: NCRG, Aston University, UK Lecturer in Information Engineering Over the next few months we will be looking to appoint 2 new lecturers and several postdoctoral research fellows to contribute to several funded international collaborative projects that the Group is involved in, in the broad areas of machine learning and system complexity issues that might be of interest to readers of this group. In the first instance, we are currently seeking to appoint an enthusiastic research-led individual as a full time lecturer to the Information Engineering Group within the School of Engineering and Applied Science who is also capable of contributing quality teaching to our undergraduate and graduate taught programmes which cover the mathematics provision of the School. The research areas of the Group are diverse, covering both theoretical and practical aspects of information analysis, with emphasis on the study of complex systems. Current interests include biomedical signal and pattern processing, bioinformatics, inference and graphical models, theory and algorithms of novel neural network structures, nonlinear dynamical modelling, data fusion, and the relationships between statistical physics and a variety of problems in information analysis, complexity and communications (specifically error correcting codes, steganography, cryptography and communication in distributed systems). Much of this work is in collaboration with other groups overseas and we are keen to extend this international scientific interaction. We are seeking someone who feels they can fit into the ethos of our Group and can contribute to these or related areas. Salary Range: 22,191 to 33,679, exceptionally to 37,629. Application forms and further particulars are available through our personnel web site: http://www.aston.ac.uk/hr/recruitment.htm, or by telephoning: (+44/0) 121 359 0870 (24 hour answerphone), or by email: b.a.power at aston.ac.uk, quoting reference number A04/61/99. Closing date for applications: 31 March 2004. Informal enquiries about the position, or indeed any of the upcoming positions, can be made to Professor David Lowe (d.lowe at aston.ac.uk), or Professor David Saad (d.saad at aston.ac.uk). From sfr at unipg.it Thu Feb 19 20:59:07 2004 From: sfr at unipg.it (sfr@unipg.it) Date: Fri, 20 Feb 2004 02:59:07 +0100 Subject: On blind deconvolution by neural systems Message-ID: <1077242347.403569eb9a80f@webmail.unipg.it> Dear Colleagues, I take the liberty to announce the availability of the following two new papers on blind deconvolution by neural systems: [1] "Analysis of Modified `Bussgang' Algorithms (MBA) for Channel Equalization" (by S. Fiori). To appear in the IEEE Trans. on Circuits and Systems - Part I Abstract: In our previous works, we introduced two modified `Bussgang' algorithms for blind channel equalization based on neural Bayesian iterative estimation of the source sequence. They were developed in order to reduce the computational complexity of the original `Bussgang' algorithm as well as to make it more flexible by introducing a kind of source adaptivity. However, the previous work relied on some heuristic findings, validated by series of computer-based experiments. The aim of this paper is to present a theoretical investigation of some particular aspects of the adapting equations, namely, the steady-state conditions, in order to ameliorate the performances of the modified `Bussgang' algorithms and to better explain their numerical behavior. [2] "A Fast Fixed-Point Neural Blind Deconvolution Algorithm" (by S. Fiori). To appear in the IEEE Trans. on Neural Networks Abstract: The aim of this contribution is to introduce a new blind deconvolution algorithm based on fixed-point optimization of a `Bussgang'-type cost function. The cost function relies on approximate Bayesian estimation achieved by an adaptive neuron. The main feature of the presented algorithm is fast convergence that guarantees good deconvolution performances with limited computational demand compared to algorithms of the same class. ** The draft versions of these contributions may be downloaded from the web-page http://www.unipg.it/sfr/ ("Publications" link) With best regards, Simone Fiori Faculty of Engineering - University of Perugia (Italy) Temporary visitor of Lab. for Mathematical Neuroscience (RIKEN, Japan) From gary.marcus at nyu.edu Sun Feb 22 13:24:18 2004 From: gary.marcus at nyu.edu (Gary Marcus) Date: Sun, 22 Feb 2004 13:24:18 -0500 Subject: The Birth of Mind [book announcement] Message-ID: Announcing [and with apologies for multiple postings] The Birth of the Mind How A Tiny Number of Genes Creates the Complexity of Human Thought By Gary Marcus "A joy to read." -- Publisher's Weekly "Expert and Lucid" -- Noam Chomsky "Brilliantly Original" -- Steven Pinker "[Across] such diverse disciplines as evolution, genetics, gene expression, cell biology, neurobiology, and psychology, Marcus .... makes the relevant issues understandable to the lay reader, and does an even better job of dispelling the myths that impede the way we think about genes and their role in making brains, and hence minds." -- Nature From the Jacket The Human Genome Project has blazed new trails in medical science and genetic research. We know that within hours of their birth, babies can recognize faces, connect what they hear with what they see and tell the difference between Dutch and Japanese. Our genes prepare us to observe the world; they shape the finest details of the human brain. But as far as psychology is concerned, writes award-winning cognitive scientist Gary Marcus, "it's almost as if Watson and Crick never met DNA." With The Birth of the Mind , Gary Marcus enters the nature vs. nurture debate and changes it forever. Genetics isn't destiny, but the only way to know what nature brings to the table, he argues, is to take a look at what genes actually do. Startling findings have recently revealed that the genome is much smaller than we once thought, containing no more than 30,000-40,000 genes. Since this discovery, scientists have struggled to understand how such a tiny number of genes could contain the instructions for building the human brain, arguably the most complex device in the known universe. Synthesizing up-to-the-minute research with his own original findings on child development, Marcus is the first to resolve this apparent contradiction as he chronicles exactly how genes create the infinite complexities of the human mind. Along the way, he reveals the common misconceptions people harbor about genes, and explores the stunning implications of this research for the future of genetic engineering. January 2004 (Basic Books). From ckiw at inf.ed.ac.uk Mon Feb 23 12:05:41 2004 From: ckiw at inf.ed.ac.uk (Chris Williams) Date: Mon, 23 Feb 2004 17:05:41 +0000 (GMT) Subject: Faculty Positions at the British Univeristy in Dubai Message-ID: [Note that machine learning is one of the areas highlighted. Please respond to the addresses given below.] British University in Dubai Institute of Informatics Chair and 3 Lectureships Applications are invited for the Post of Professor of Informatics and Director of the Institute, together with three lectureships in the new British University in Dubai. The British University in Dubai is a unique development in higher education in the Middle East, providing cutting edge research and education in key areas of science and technology. Each of the individual disciplinary areas of the University is strongly associated with and supported by 5/5* rated departments in leading research-intensive UK universities. The Institute of Informatics is associated with Edinburgh University and its world ranking School of Informatics. The new Professor and Director of the Institute will provide leadership in creating innovative research and teaching programmes and be involved in appointments to the lectureships. Appointment to the chair and the 3 lectureships will be made in areas of Informatics related to the first programmes to be developed by the Institute in: Natural Language Engineering, Knowledge Management & Engineering, and Machine Learning. Early research and teaching programmes will be developed in association with the University of Edinburgh's 5* rated School of Informatics. Newly appointed staff will spend part of their first year working with colleagues in Edinburgh and will be eligible for Honorary Fellowships in the University of Edinburgh. Applicants for the chair must have a strong research record. Appointment will be made on the Professorial scale of a minimum of AED 25,000 per month. Applicants for lectureships should have a PhD or equivalent, a strong and growing publication list and evidence of excellent potential in research and will be appointed on the scale AED 18,000 to AED 28,000 per month. All appointees will be eligible for a remuneration package including housing, medical, flights and set-up package. Further particulars of the posts and details of the application process can be obtained from either www.buid.ac.ae or Professor G.S.Boulton, Vice Principal, University of Edinburgh; email: Francesca.Casci at ed.ac.uk. The closing date for applications is 19 March. Interviews will be held in late March/early April. From knoblauch at neuro.informatik.uni-ulm.de Mon Feb 23 11:44:08 2004 From: knoblauch at neuro.informatik.uni-ulm.de (Andreas Knoblauch) Date: Mon, 23 Feb 2004 17:44:08 +0100 Subject: PhD thesis available: Spiking associative memory and modeling visual areas Message-ID: Dear Connectionists, My PhD thesis on "Synchronization and pattern separation in spiking associative memories and visual cortical areas" is now available on the WWW at http://www.informatik.uni-ulm.de/ni/publ/AKnoblauch/Knoblauch2003B.pdf http://www.informatik.uni-ulm.de/ni/mitarbeiter/AKnoblauch.html For those interested, I have appended the table of contents Best regards, Andreas Knoblauch. ======================================================= Keywords: associative memory, cell assemblies, spiking neural networks, spike time coding, spike synchronization, sparse distributed coding, feature binding, scene segmentation, visual cortex. table of contents: 1 Overview 1.1 Motivation and classification 1.2 Organization 2 Introduction 2.1 Neurobiology of the brain 2.2 Distributed representations, neural assemblies, and binding 2.3 Associative memory 2.4 Simulation tools and models of neurons and synapses 3 Willshaw associative memory 3.1 Classical analysis of the Willshaw model 3.2 Matrix compression yields storage capacity 1 3.3 Fault tolerance 3.4 Retrieval efficiency for sequential implementations 3.5 Retrieval efficiency for parallel implementations 3.6 A critique of the classical binomial analysis 3.7 Implications 4 Spiking associative memory 4.1 Addressing with superpositions 4.2 Pattern separation 4.3 Spike counter model 4.4 Analysis of the spike counter model 4.5 Technical implementations of the spike counter model 4.6 Biological implementations of the spike counter model 4.7 Implications 5 Model of visual cortical areas 5.1 Modeling 5.2 Interaction between two cortical areas 5.3 Relation to neurophysiological results 5.4 Summary of the mechanisms in the model 6 Binding in the brain 6.1 Is there a binding problem in the brain? 6.2 Critique of the strong Temporal Correlation Hypothesis 6.3 Sketch of a more global model of binding in cortical areas 7 Conclusions 7.1 Contributions to the theory of Willshaw associative memory 7.2 Contributions to the theory of cell assemblies 8 Zusammenfassung (in German) Appendix A Information theory A.1 Basic information theory A.2 Binary channels A.3 Optimal compression of sparse bit streams B Analysis of associative memory B.1 Information-theoretical storage capacity B.2 Auto association and hetero association B.3 Retrieval and separation quality B.4 Further analysis of Willshaw associative memory C Simulation tool Felix++ C.1 Basic architecture of Felix++ C.2 Simulation environment and components: Code examples C.3 Structure of a Felix++ simulation D Symbol reference -- --------------------------------------------------------------------------------------------------- Andreas Knoblauch Abt. Neuroinformatik Universitt Ulm James-Franck-Ring D-89069 Ulm Tel. : ++49-731-502-4255 Fax : ++49-731-502-4156 email : knoblauch at neuro.informatik.uni-ulm.de WWW : http://www.informatik.uni-ulm.de/ni/mitarbeiter/AKnoblauch.html --------------------------------------------------------------------------------------------------- From nnk at atr.co.jp Tue Feb 24 04:27:51 2004 From: nnk at atr.co.jp (Neural Networks Editorial Office) Date: Tue, 24 Feb 2004 18:27:51 +0900 Subject: Neural Networks 17(2) Message-ID: NEURAL NETWORKS 17(2) Contents - Volume 17, Number 2 - 2004 ------------------------------------------------------------------ *** NEURAL NETWORKS LETTERS *** "A computational model of spatio-temporal dynamics in depth filling-in" Shigeaki Nishina, Mitsuo Kawato "Predicting spike timings of current-injected neurons" Yasuhiro Tsuboa, Takeshi Kaneko, Shigeru Shinomoto ***** Neuroscience and Neuropsychology ***** "Configural representations in transverse patterning with a hippocampal model" Paul Rodriguez, William B. Levy "Clustering behavior in a three-layer system mimicking olivo-cerebellar dynamics" Manuel G. Velarde, Vladimir I. Nekorkin, Valeri A. Makarov, Vladimir I. Makarenko, Rodolfo R. Llina?s ***** Mathematical and Computational Analysis ***** "Information processing in a neuron ensemble with the multiplicative correlation structure" Si Wu, Shun-ichi Amari, Hiroyuki Nakahara "On-line learning through simple perceptron learning with a margin" Kazuyuki Hara, Masato Okada "Bifurcating neuron: computation and learning" Mykola Lysetskiy, Jacek M. Zuradacor "Generalized neural networks for spectral analysis: dynamics and Liapunov functions" Jose? M. Vegas, Pedro J. Zufiria "A principal components analysis self-organizing map" Ezequiel Lo?pez-Rubio, Jose? Mun??oz-Pe?rez, Jose? Antonio Go?mez-Ruiz "Regularising neural networks using flexible multivariate activation function" Mirko Solazzi, Aurelio Uncini "Neural networks for nonlinear and mixed complementarity problems and their applications" Chuangyin Dang, Yee Leung, Xing-bao Gao, Kai-zhou Chen "A canonical correlation neural network for multicollinearity and functional data" Zhenkun Gou, Colin Fyfe CURRENT EVENTS ------------------------------------------------------------------ Electronic access: www.elsevier.com/locate/neunet/. Individuals can look up instructions, aims & scope, see news, tables of contents, etc. Those who are at institutions which subscribe to Neural Networks get access to full article text as part of the institutional subscription. Sample copies can be requested for free and back issues can be ordered through the Elsevier customer support offices: nlinfo-f at elsevier.nl usinfo-f at elsevier.com or info at elsevier.co.jp ------------------------------ INNS/ENNS/JNNS Membership includes a subscription to Neural Networks: The International (INNS), European (ENNS), and Japanese (JNNS) Neural Network Societies are associations of scientists, engineers, students, and others seeking to learn about and advance the understanding of the modeling of behavioral and brain processes, and the application of neural modeling concepts to technological problems. Membership in any of the societies includes a subscription to Neural Networks, the official journal of the societies. Application forms should be sent to all the societies you want to apply to (for example, one as a member with subscription and the other one or two as a member without subscription). The JNNS does not accept credit cards or checks; to apply to the JNNS, send in the application form and wait for instructions about remitting payment. The ENNS accepts bank orders in Swedish Crowns (SEK) or credit cards. The INNS does not invoice for payment. ---------------------------------------------------------------------------- Membership Type INNS ENNS JNNS ---------------------------------------------------------------------------- membership with $80 (regular) SEK 660 Y 13,000 Neural Networks (plus Y 2,000 enrollment fee) $20 (student) SEK 460 Y11,000 (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- membership without $30 SEK 200 not available to Neural Networks non-students (subscribe through another society) Y 5,000 student (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- Name: _____________________________________ Title: _____________________________________ Address: _____________________________________ _____________________________________ _____________________________________ Phone: _____________________________________ Fax: _____________________________________ Email: _____________________________________ Payment: [ ] Check or money order enclosed, payable to INNS or ENNS OR [ ] Charge my VISA or MasterCard card number ____________________________ expiration date ________________________ INNS Membership 19 Mantua Road Mount Royal NJ 08061 USA 856 423 0162 (phone) 856 423 3420 (fax) innshq at talley.com http://www.inns.org ENNS Membership University of Skovde P.O. Box 408 531 28 Skovde Sweden 46 500 44 83 37 (phone) 46 500 44 83 99 (fax) enns at ida.his.se http://www.his.se/ida/enns JNNS Membership c/o Professor Shozo Yasui Kyushu Institute of Technology Graduate School of Life Science and Engineering 2-4 Hibikino, Wakamatsu-ku Kitakyushu 808-0196 Japan 81 93 695 6108 (phone and fax) jnns at brain.kyutech.ac.jp http://www.jnns.org/ ---------------------------------------------------------------------------- From mseeger at EECS.berkeley.EDU Tue Feb 24 14:23:21 2004 From: mseeger at EECS.berkeley.EDU (Matthias Seeger) Date: Tue, 24 Feb 2004 11:23:21 -0800 (PST) Subject: Gaussian Processes for Machine Learning Message-ID: Dear colleagues, the tutorial paper Gaussian Processes for Machine Learning will appear in the International Journal for Neural Systems (IJNS) and is available from www.cs.berkeley.edu/~mseeger/papers/bayesgp-tut.html. I hope this is interesting to some of you. Best wishes, Matthias ---- Abstract: Gaussian processes (GPs) are natural generalisations of multivariate Gaussian random variables to infinite (countably or continuous) index sets. GPs have been applied in a large number of fields to a diverse range of ends, and very many deep theoretical analyses of various properties are available. This paper gives an introduction to Gaussian processes on a fairly elementary level with special emphasis on characteristics relevant in machine learning. It draws explicit connections to branches such as spline smoothing models and support vector machines in which similar ideas have been investigated. --- Matthias Seeger Tel: 510-642-8468 485 Soda Hall, UC Berkeley Fax: 510-642-5775 Berkeley, CA 94720-1776 www.cs.berkeley.edu/~mseeger From thomas.shultz at mcgill.ca Tue Feb 24 11:55:29 2004 From: thomas.shultz at mcgill.ca (Thomas Shultz) Date: Tue, 24 Feb 2004 11:55:29 -0500 Subject: book announcement Message-ID: <000c01c3faf7$057dc320$b86ace84@psych.mcgill.ca> This book may be of interest to readers of Connectionists Computational Developmental Psychology by Thomas R. Shultz Cambridge, MA: MIT Press, 2003 $38 322 pages ISBN 0-262-19483-X Despite decades of scientific research, the core issues of child development remain too complex to be explained by traditional verbal theories. These issues include structure and transition, representation and processing, innate and experiential determinants of development, and stages of development. This book shows how computational modeling can be used to capture these complex phenomena, and lays the foundation for a new subfield of developmental psychology, computational developmental psychology. The book compares rule-based and connectionist modeling, and features constructivist models employing networks that evolve as well as learn. This allows for qualitative changes in processing mechanisms and a way to implement synaptogenesis and neurogenesis. The appendices provide additional background on the mathematical concepts used, and the book's website contains easy-to-use computational packages. Thomas R. Shultz is Professor of Psychology and Computer Science and Director of the Laboratory of Natural and Simulated Cognition at McGill University. ----------------------------------------------------------------- Thomas Shultz, Professor, Department of Psychology McGill University, 1205 Penfield Ave., Montreal, Quebec,=20 Canada H3A 1B1. Associate Member, School of Computer Science E-mail: thomas.shultz at mcgill.ca Updated 20 February 2004: http://www.psych.mcgill.ca/perpg/fac/shultz/personal/default.htm Phone: 514 398-6139 Fax: 514 398-4896 ----------------------------------------------------------------- From schunn+ at pitt.edu Wed Feb 25 11:08:36 2004 From: schunn+ at pitt.edu (Christian Schunn) Date: Wed, 25 Feb 2004 11:08:36 -0500 Subject: doctoral consortium for ICCM2004 Message-ID: NEW ANNOUNCEMENT FOR ICCM2004 The 6th International Conference on Cognitive Modeling announces a (funded) opportunity for doctoral students to attend a doctoral consortium, a small 1-day workshop just prior to the conference. The ICCM 2004 Doctoral Consortium provides an opportunity for doctoral students to explore their research interests in a multi-approach workshop, under the guidance of a panel of distinguished research faculty, consisting of David Touretzky (CMU) and Wayne Gray (RPI). The Consortium has the following objectives: * provide a setting for mutual feedback on participants' current research and guidance on future research directions * develop a supportive community of scholars and a spirit of collaborative research * contribute to the conference goals through interaction with other researchers and conference events The Consortium will be held the day before ICCM-2004, on Wednesday, July 29th, 2004. Student participants receive complimentary conference registration as well as lodging and travel reimbursement-maximum amounts will be determined at a later date. In addition to participating in the Doctoral Consortium, all Doctoral Consortium students are expected to prepare a poster for presentation at the ICCM-2004 poster session. Deadline: April 1st 2004 Review Criteria The Doctoral Consortium review committee will select participants based on their anticipated contribution to the Consortium objectives. Participants typically have settled on thesis directions but have not necessarily had their research proposals accepted by their thesis committees. The Doctoral Consortium encourages participation of students from a wide variety of modeling approaches. Students from under-represented groups or institutions, including students from institutions where modeling is not a strength, are especially encouraged to apply. Format To apply, prepare and email to iccm at pitt.edu the following: an extended abstract in PDF form and a letter of recommendation. In the email sent to iccm at pitt.edu, include the following information: your name, your department, your institution, your thesis advisor's name, and keywords that describe your research area (both modeling approach and domain area). Extended Abstract Prepare a two-page extended abstract of your thesis work in the Conference Publications Format including title, author information, abstract, keywords, thesis research summary, and references. The abstract will appear in the proceedings. Submit the abstract as a PDF file by email attachment to iccm at pitt.edu. Put your name and "Doctoral Consortium" in the email message subject field. In the email message, include your name, your department, your institution, your thesis advisor's name, and keywords that describe your research area (both modeling approach and domain area). Letter of Recommendation Have a letter of recommendation from your thesis advisor emailed to iccm at pitt.edu (by April 1st 2004). The recommendation should include formal reviews of your research and the expected timetable for completion. Posters Doctoral Consortium students are not expected to do a separate poster submission for the work that they plan to discuss at the Consortium meeting. (A submission on other work is, of course, acceptable.) Upon Acceptance Applicants will be notified by April 4th 2004. Extended abstracts of accepted submissions will be published in the ICCM 2004 Proceedings. Author Kits with instructions on how to submit materials for publication will be sent with acceptance. At the Conference Authors must be available for discussion and interaction with attendees during the scheduled poster session. Submissions * Your submission must be in English. * Submissions arriving after the deadline will not be considered. * You will receive email notification upon receipt of your submission. For more information on ICCM2004 more generally, see http://simon.lrdc.pitt.edu/~iccm/ From bressler at fau.edu Tue Feb 24 13:05:05 2004 From: bressler at fau.edu (Steven L Bressler) Date: Tue, 24 Feb 2004 13:05:05 -0500 Subject: POSTDOCTORAL POSITION AVAILABLE Message-ID: POSTDOCTORAL POSITION AVAILABLE Center for Complex Systems & Brain Sciences Florida Atlantic University A postdoctoral position is open immediately in the laboratory of Dr. Steven Bressler ( http://www.ccs.fau.edu/~bressler/ ) at the Center for Complex Systems & Brain Sciences at Florida Atlantic University ( http://www.ccs.fau.edu/ ) to participate in an ongoing research project studying the effects of attention on visuomotor processing in the cerebral cortex. The project is funded by research grants from the National Science Foundation and the National Institute of Mental Health. Desired background: -- Experience in time series analysis and multivariate statistics -- Expertise in MATLAB and/or C programming -- Experience in experimental design Please send curriculum vitae, illustrative research articles, and the names and e-mail addresses of three references to Dr. Steven Bressler at bressler at fau.edu . Steven L. Bressler, Ph.D. Professor Complex Systems & Brain Sciences Bressler Home Page From G.Brown at cs.bham.ac.uk Tue Feb 24 11:23:09 2004 From: G.Brown at cs.bham.ac.uk (Gavin Brown) Date: Tue, 24 Feb 2004 16:23:09 +0000 (GMT) Subject: PhD available: Diversity in Neural Network Ensembles In-Reply-To: References: Message-ID: Dear Connectionists, My PhD thesis on "Diversity in Neural Network Ensembles" is now available on the web at: http://www.cs.bham.ac.uk/~gxb/research/gbrown_thesis.ps.gz Abstract: --------------------------------------- We study the issue of error diversity in ensembles of neural networks. In ensembles of regression estimators, the \emph{measurement} of diversity can be formalised as the Bias-Variance-Covariance decomposition. In ensembles of classifiers, there is no neat theory in the literature to date. Our objective is to understand how to precisely define, measure, and create diverse errors for both cases. As a focal point we study one algorithm, \emph{Negative Correlation (NC) Learning} which claimed, and showed empirical evidence, to \emph{enforce} useful error diversity, creating neural network ensembles with very competitive performance on both classification and regression problems. With the lack of a solid understanding of its dynamics, we engage in a theoretical and empirical investigation. During the theoretical investigations, we find that NC succeeds due to exploiting the \emph{Ambiguity decomposition} of mean squared error. We provide a grounding for NC in a statistics context of bias, variance and covariance, including a link to a number of other algorithms that have exploited Ambiguity. The discoveries we make regarding NC are not limited to neural networks. The majority of observations we make are in fact properties of the mean squared error function. We find that NC is therefore best viewed as a \emph{framework}, rather than an algorithm itself, meaning several other learning techniques could make use of it. We further study the configurable parameter in NC, thought to be entirely problem-dependent, and find that one part of it can be analytically determined for any ensemble architecture. We proceed to define an upper bound on the remaining part of the parameter, and show considerable empirical evidence that a lower bound also exists. As the size of the ensemble increases, the upper and lower bounds converge, indicating that the optimal parameter can be determined exactly. We describe a number of experiments with different datasets and ensemble architectures, including the first comparisons to other popular ensemble methods; we find NC to be a competitive technique, worthy of further application. Finally we conclude with observations on how this investigation has impacted our understanding of diversity in general, and note several possible new directions that are suggested by this work. This includes links to evolutionary computation, Mixtures of Experts, and regularisation techniques. --------------------------------------------- Enjoy, -Gav From dalche at epigenomique.org Thu Feb 26 05:27:51 2004 From: dalche at epigenomique.org (Florence Dalche) Date: Thu, 26 Feb 2004 11:27:51 +0100 Subject: call for challenge proposals Message-ID: <403DCA27.7050902@epigenomique.org> ******************************************************************************************** 2nd CALL FOR CHALLENGE PROPOSALS of PASCAL european network of excellence (Pattern Analysis, Statistical modelling and ComputationAl Learning) http://www.pascal-network.org/challenge DEADLINE : March 4th, 2004 ******************************************************************************************** The PASCAL network of excellence invites submissions of challenge proposals for which the PASCAL joint program of challenges will provide funding and support to cover part of expenses. The aim of this challenge program is to focus attention of the research community on difficult taks that may require new algorithmic development. Both application challenge and theoretical challenges can be proposed. For application challenges,the idea is to run competitions to get no-bullshit head-to-head comparisons of established or new techniques on well chosen datasets. We expect that results will provide interesting information to a research field about comparative performance of different methods under controlled conditions. Canonical examples are the NIPS feature selection challenge (2003), the various NIST speech competitions. The proposed challenge can be a concrete task in the following fields of application of machine learning related to PASCAL: Machine vision Speech Haptics Brain computer interface Natural language processing Information Retrieval/ Textual Information Access Text-mining and understanding Multimodal integration User modelling for computer human interaction We also invite submisssions about theoretical questions and open problems relevant to pattern analysis,statistical modelling and computational learning.In this case, evaluation rules can differ and can be stated as a serie of steps to be reached. Once a proposal will be accepted, the challenge submitter will have in charge the organization and the tracking of the challenge with the help and support of the challenge program in link with the infrastructure program. Results will be presented at a challenge workshop organized at the end of the challenges. The first workshop challenge will be held in November 2004 with others evry 6 months. Web site: ======== A web site dedicated to the challenge submission process is now opened on http://www.pascal-network.org/challenge. Submission: ========== Authors of challenge proposals will be asked to answer the following questions : * Title of the proposal * Description of the challenge (including the major problem to be addressed) * Relevance with PASCAL * Expected impact on machine learning research and field of interest * Feasibility * Description/ availability of the data * Time and budget required to set up the datasets and resources * Evaluation procedure * Results already obtained (established baseline) * Estimated number of teams interested in the participation of the challenge * Indication of short term/long term challenge Key dates : ========= DEADLINE for electronic submissions : March 4th Notification to the authors : March 12th Reception of corrected electronic submissions and public notification of selected proposals: March 19th Reviewing process: ================= Each submitted challenge proposal would be reviewed by two members of the programme committee. A detailed feedback will be send back to the authors. When proposals will be accepted, there will be a short time for adjusting it to answer the comments of the reviewers from March 12th to March 19th. Preliminary programme committee: =============================== Samy Bengio (IDIAP, Martigny) Walter Daelemans (University of Antwerp) Cyril Goutte (Xerox, Grenoble) Steve Gunn (University of Southampton) Klaus-Robert Mueller (Faunhofer,Berlin) Michle Sebag (LRI, Orsay) Bill Triggs (INRIA, Grenoble) Chris Watkins (Royal Holloway university, London) Florence d'Alch-Buc (university of Paris 6) Any question regarding this call for challenge proposals may be addressed to Florence d'Alch-Buc : florence.dalche at lip6.fr with PASCAL CHALLENGE in the subject. *************************************************************************************** From frankmj at grey.colorado.edu Thu Feb 26 12:22:31 2004 From: frankmj at grey.colorado.edu (Michael J. Frank) Date: Thu, 26 Feb 2004 10:22:31 -0700 (MST) Subject: Computational Model of Cognitive Deficits in Medicated and Non-medicated Parkinsonism. Message-ID: Readers of this list may be interested in the following preprint, to appear in the Journal of Cognitive Neuroscience: http://psych.colorado.edu/~frankmj/papers/BG_DA_model.preprint.pdf Frank, M.J. (in press). Dynamic dopamine modulation in the basal ganglia: A neurocomputational account of cognitive deficits in medicated and non-medicated Parkinsonism. Journal of Cognitive Neuroscience. Abstract: Dopamine (DA) depletion in the basal ganglia (BG) of Parkinson's patients gives rise to both frontal-like and implicit learning impairments. Dopaminergic medication alleviates some cognitive deficits but impairs those that depend on intact areas of the BG, apparently due to DA ``overdose''. These findings are difficult to accommodate with verbal theories of BG/DA function, owing to complexity of system dynamics: DA dynamically modulates function in the BG, which is itself a modulatory system. This paper presents a neural network model that instantiates key biological properties and provides insight into the underlying role of DA in the BG during learning and execution of cognitive tasks. Specifically, the BG modulates the execution of ``actions'' (e.g., motor responses and working memory updating) that are being considered in different parts of frontal cortex. Phasic changes in DA, which occur during error feedback, dynamically modulate the BG threshold for facilitating/suppressing a cortical command in response to particular stimuli. Reduced dynamic range of DA explains Parkinson and DA overdose deficits with a single underlying dysfunction, despite overall differences in raw DA levels. Simulated Parkinsonism and medication effects provide a theoretical basis for behavioral data in probabilistic classification and reversal tasks. The model also provides novel testable predictions for neuropsychological and pharmacological studies, and motivates further investigation of BG/DA interactions with prefrontal cortex in working memory. From joseph at nats.informatik.uni-hamburg.de Fri Feb 27 03:23:24 2004 From: joseph at nats.informatik.uni-hamburg.de (Joseph Chen) Date: Fri, 27 Feb 2004 09:23:24 +0100 Subject: PhD Thesis on "Quantum Computation and Natural Language Processing" Message-ID: <361E265E-68FE-11D8-BC1C-003065C8C1CE@nats.informatik.uni-hamburg.de> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Dear Connectionists, The PhD thesis: "Quantum Computation and Natural Language Processing" is available at http://www.sub.uni-hamburg.de/opus/volltexte/2002/769/pdf/ dissertation.pdf This may be of interest to some of you. Best Regards, Joseph Chen ====== Abstract ====== In this thesis, a novel approach to natural language understanding inspired by quantum mechanical principle is proposed. It is based on an analogy between the physical objects at the quantum level and human's mental states. In this way, the physical and the mental phenomena are to be understood within the same framework. It is also proposed that the apparent differences between mind and matter do not lie in the fundamental differences of their properties, but in the different manifestation of macroscopic matter and macroscopic mind owing to their different composition of pure quantum eigenstates. The apparent differences are therefore quantitative rather than qualitative. Specifically, symbols in various cognitive functions are to be treated as eigenstates with respect to a particular quantum experimental arrangement. Moreover, I claim that reasoning and inference can be treated as transformations of semiosis with symbols being the eigenstates of a particular formulation operator. The operator is the counterpart of an observable in quantum mechanics. A state of affairs (a superposition of these eigenstates) does not have well-defined physical properties until it is actually measured. Consequently the classical semantics (as classical symbols' referring to the classical physical reality) is also not well-defined and may be a misleading idea. Different from classical semantics, meaning in the quantum mechanical framework should be treated as an active measurement done on a state of affair. Moreover, the ill-definedness also manifests itself in the cognition internal to a person if we regard memory as a language-like representational system. Nevertheless, memory, treated as a specific language system, is a largely quasi-classical phenomenon in that the chemical activities in the brain are an aggregate limiting case of quantum mechanics with a very large number of quanta. The classical ``objective'' physical reality is therefore a limiting case of quantum reality as well. The general language in which common sense logic is embedded is then investigated and the apparent evasiveness and ambiguity of language can be accommodated in a quantum framework. This is done by postulating an analogous Uncertainty Principle and observing the implication of it. An important implication is the ``concept-symbol'' duality. As applications, the quantum mechanical formalism is applied to cognitive processes. For instance, non-monotonicity and counterfactual conditionals can be accommodated and assimilated in this framework. Specifically, the time-asymmetric property and the genuine unknown state of non-monotonic reasoning can be easily explained in quantum mechanics. This is also the case for the potentiality and actuality, which are crucial ideas for explaining counterfactual reasoning. Furthermore, causality can be regarded as a disguise of counterfactual reasoning. The second part of the thesis is devoted to simulations and technical applications of the aforementioned principle in natural language processing. First the preliminary experiments of common sense logic are presented. These show that the ``classicization'' of common sense logic can be implemented with very simple quantum mechanical systems. Moreover, the richness of the quantum framework goes well beyond what a classical system can offer. There can be ``fine-structures'' within seemingly simple logical arguments (XOR, for example). This is also the case for non-monotonic and counterfactual reasoning. Simple natural language tasks are also simulated based on different natural language corpora. First the syllogistic arguments embedded in natural language are simulated with a quantum system, which delivers quite remarkable results. Secondly, a monolingual syntax manipulation is implemented with a quantum system, in which the quantum mechanical approach can achieve much better performance than connectionist one. In the last experiment, a quantum mechanical architecture is trained for bilingual translation between English and German, in which there are several thorny properties in the natural language corpus, for example lexical ambiguity, separable prefixes, complicated conjugation, and non-linear translational word mappings. Nevertheless, the quantum mechanic architecture can deliver very satisfactory results. - -- Dr. Joseph C.H. Chen Computer Science Department, University of Hamburg Vogt-Koelln-Str. 30 22527 Hamburg http://nats-www.informatik.uni-hamburg.de/~joseph/ +49-40-42883-2523 (O) +49-40-42883-2515 (FAX) PGP Public Key: http://nats-www.informatik.uni-hamburg.de/~joseph/joseph.pgp.asc Printed on Planet Earth and with 100% recycled electrons. - ---------------------------------------------------------------- -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (Darwin) iD8DBQFAPv6ANE0t/Ns3SDgRAtPlAJ9NMIfBPlY+aLRYnD4xyxqcp5xdUQCeO1hK NOwwv80fRi4TcJtqfdHIxQ0= =XcFl -----END PGP SIGNATURE----- From rich at ini.phys.ethz.ch Fri Feb 27 14:55:36 2004 From: rich at ini.phys.ethz.ch (Richard Hahnloser) Date: Fri, 27 Feb 2004 11:55:36 -0800 Subject: PhD Position available at the ETH in Zurich Message-ID: <403FA0B8.7030401@ini.phys.ethz.ch> Applications are invited for the position of a Doctoral Student (PhD student) to investigate novel methods of online and offline analysis of electrophysiological and acoustic data. We are a well funded research group studying the neural mechanisms of learned vocalizations in the songbird. For more information about the lab, visit http://www.ini.unizh.ch/~rich. We are seeking a motivated person with a strong background in physics and / or computer science. Research topics include but are not limited to learning algorithms for real-time song recognition and statistical methods for spike-train analysis. It will be expected of the person to closely collaborate with experimentalists in the lab. Position is available immediately for individuals with a degree in related fields. Salary is according to the guidelines of the Swiss National Science Foundation. The Institute of Neuroinformatics is a multidisciplinary institution affiliated with the Department of Physics at the Swiss Federal Institute of Technology and is located on the Irchel Campus of the University of Zurich, in the city of Zurich. For more information about the institute, visit http://www.ini.unizh.ch. Interested applicants should direct inquiry or applications with a CV and names of two references to: Dr. Richard Hahnloser Institute for Neuroinformatics, UNIZH/ETHZ Winterthurerstrasse 190 8057 Zurich Switzerland Tel: +41 1 635 3060 Email: rich at ini.phys.ethz.ch