From silvia at sa.infn.it Mon Aug 1 08:56:43 2005 From: silvia at sa.infn.it (Silvia Scarpetta) Date: Mon, 01 Aug 2005 14:56:43 +0200 Subject: Connectionists: opening position in neural computation in Salerno Message-ID: OPENING A two-year postdoctoral position (Assegno di Ricerca) in Neural Networks / NeuroPhysics starting on next Fall 2005 (around October, 2005) is available at the Dept. of Physics "E.R. Caianiello" of the Universit degli Studi di Salerno, Italy, in the group headed by Prof. Maria Marinaro (http://www.sa.infn.it/NeuralGroup/) The monthly net amount of the fellowship is about 1200 Euro. RESEARCH ACTIVITY Areas of particular interest include: 1) Neurobiological applications of theoretical physics tools, computational and mathematical modeling of neural dynamics, cortical dynamics and oscillations, learning memory and modelling of STDP, rhythmic locomotion and related topics. 2) Neural networks applied to signal and image processing and to speech recognition. Description of current project in our group can be found in: http://www.sa.infn.it/NeuralGroup/ MINIMUM REQUIREMENTS master degree in scientific disciplines (physics, neuroscience, mathematics, computer science, engeneering, etc) and a PhD degree or a 3 years research experience in fields related with the position topic. Sufficient knowledge of the English language and of a programming language (C or Matlab) will be required for successful project work. *** Provide, please, a CV with publication list and description of research interests, and arrange for at least two letters of reference to be sent to: Prof. Maria Marinaro Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende I-84081 Baronissi (SA) Italy ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Prof. Maria Marinaro Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende 84081 Baronissi (SA) Italy Tel. +39 089 965318 Web page: http://www.sa.infn.it/NeuralGroup/ Dr. Silvia Scarpetta Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende 84081 Baronissi (SA) Italy Tel. +39 089 965318 Web page: http://www.sa.infn.it/silvia.scarpetta From marc.vanhulle at med.kuleuven.ac.be Tue Aug 2 07:34:02 2005 From: marc.vanhulle at med.kuleuven.ac.be (Marc Van Hulle) Date: Tue, 02 Aug 2005 13:34:02 +0200 Subject: Connectionists: differential log-likelihood Message-ID: <42EF5A2A.ABD44E7B@med.kuleuven.ac.be> Dear Connectionists, Two papers on a new and unbiased information metric, called differential log-likelihood, are available. Regular information metrics such as Akaike's, BIC and GIC start from the log-likelihood but do not produce an unbiased metric. This implies that the optimal case corresponds to a minimum. The differential log-likelihood is optimal when it reaches the zero value, and a zero-crossing is much easier to locate than a minimum. Van Hulle, M.M. (2005). Differential log-likelihood for evaluating and learning Gaussian mixtures. Neural Computation, in press. Abstract We introduce a new unbiased metric for assessing the quality of density estimation based on Gaussian mixtures, called differential log-likelihood. As an application, we determine the optimal smoothness and the optimal number of kernels in Gaussian mixtures. Furthermore, we suggest a learning strategy for Gaussian mixture density estimation, and compare its performance with log-likelihood maximization for a wide range of real-world data sets. http://134.58.34.60/~marc/NC11.pdf -- Van Hulle, M.M. (2005). Mixture density modeling, Kullback-Leibler divergence, and differential log-likelihood. Signal Processing (Special Issue on Information Theoretic Signal Processing), J. Principe & D. Erdogmus (Eds.), 85(5), 951-963. http://134.58.34.60/~marc/SP.pdf -- Marc M. Van Hulle K.U.Leuven Laboratorium voor Neurofysiologie Faculteit Geneeskunde (Medical School) Campus Gasthuisberg Bus 801 Herestraat 49 B-3000 Leuven Belgium Phone: + 32 16 345961 Fax: + 32 16 345960 E-mail: marc.vanhulle at med.kuleuven.ac.be URL: http://simone.neuro.kuleuven.ac.be From terry at salk.edu Wed Aug 3 16:58:24 2005 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 03 Aug 2005 13:58:24 -0700 Subject: Connectionists: NEURAL COMPUTATION 17:9 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 17, Number 9 - September 1, 2005 REVIEW The Cocktail Party Problem Simon Haykin and Zhe Chen NOTES Edgeworth Approximation of Multivariate Differential Entropy Marc M. Van Hulle Perfect Fault Tolerance of the /n-k-n/ Network Elko B. Tchernev, Rory G. Mulvaney and Dhananjay S. Phatak On the Slow Convergence of EM and VBEM in Low-Noise Linear Models Kaare Brandt Petersen, Ole Winther and Lars Kai Hansen LETTERS Analyzing Functional Connectivity Using a Network Likelihood Model of Ensemble Neural Spiking Activity Murat Okatan, Matthew A. Wilson, and Emery N. Brown Data-Robust Tight Lower Bounds to the Information Carried by Spike Times of a Neuronal Population G. Pola, R. S. Petersen, A. Thiele, M. P. Young and S. Panzeri Fluctuation-Disspation Theorem and Models of Learning Ilya Nemenman Correlated Firing in a Feedforward Network with Mexican-Hat-Type Connectivity Kosuke Hamaguchi, Masato Okada, Michiko Yamana, and Kazuyuki Aihara Supervised Learning in a Recurrent Network of Rate-Model Neurons Exhibiting Frequency Adaptation Pierre A. Fortier, Emmanuel Guigon and Yves Burnod Learning Bounds for Kernel Regression Using Effective Data Dimensionality Tong Zhang ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2005 - VOLUME 17 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $114 $54 $57.78 Individual $100 $107.00 $143 $90 $96.30 Institution $680 $727.60 $734 $612 $654.84 * 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 wduch at phys.uni.torun.pl Fri Aug 5 20:42:39 2005 From: wduch at phys.uni.torun.pl (wduch@phys.uni.torun.pl) Date: Sat, 06 Aug 2005 02:42:39 +0200 Subject: Connectionists: Free tutorials at the ICANN conference, Sept. 11, Warsaw, Poland Message-ID: <1123288959.42f4077f1dc4c@poczta.umk.pl> We are pleased to announce the following tutorials, free for all registered participants of the ICANN-2005, http://www.ibspan.waw.pl/ICANN-2005/index.html 1. An Introduction to Bayesian Methods, by Christopher Bishop (Microsoft, Cambridge, UK) 2. Spiking neural networks, by Hans-Heinrich Bothe (Technical University of Denmark) 3. Evolving Connectionist Systems: Biological Principles, Models and Applications by Nik Kasabov (Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, New Zealand) 4. Networks of Evolutionary Processors: A Survey by Carlos Martin-Vide (Rovira i Virgili University, Spain) 5. Recurrent neural networks: state of the art, by Juergen Schmidhuber (IDSIA, Switzerland and TU Munich, Germany) 6. Through Attention to Creating Conscious Machines, by John Gerald Taylor (King's College London, UK) They will be presented on September 11, 2005 (Sunday) at the Gromada Airport Hotel, Warsaw, Poland (the conference venue). From auke.ijspeert at epfl.ch Tue Aug 9 10:52:34 2005 From: auke.ijspeert at epfl.ch (Auke Ijspeert) Date: Tue, 09 Aug 2005 16:52:34 +0200 Subject: Connectionists: AMAM2005: call for participation and last-minute-results abstracts Message-ID: <42F8C332.2070708@epfl.ch> AMAM2005, Sept 25-30 2005, Ilmenau, http://www.tu-ilmenau.de/amam Dear Connectionist colleagues, Researchers working on any aspect related to the adaptive control of movement and locomotion in animals and robots might be interested in participating to AMAM2005, the Third International Symposium on Adaptive Motion in Animals and Machines. The symposium will take place at the Technische Universit?t Ilmenau, Germany, from September 25th to September 30th, 2005. The two previous symposia in Montreal and Kyoto, which brought together researchers in neurobiology, biomechanics, neural computation, and robotics, were very exciting and fruitful events. The program is now fixed, see http://www.tu-ilmenau.de/amam , but there is the opportunity to submit an abstract to a special *"last-minute-results" poster session*, see the CFP below, and* to bring robots to participate to public robot demonstrations*, see http://www.tu-ilmenau.de/amam (Robot Data Sheet link). To hold keynotes ("Basics & Advances in ...") we have confirmations from the following colleagues: Biology & Mechanics: *Roy Ritzmann*, Movement Through Complex Terrain: From Animals to Robots and Back I *Martin S. Fischer*, Movement Through Complex Terrain: From Animals to Robots and Back II Biology & Control: *Sten Grillner*, The Neural control of vertebrate locomotion - from ion channels to behavior *Auke Jan Ijspeert*, Pattern generators in the central nervous system: numerical models and applications to robotics Bionics/Biomimetics: *Werner Nachtigall*, Technical Biology - Basis for Bionical Realization (supported by BioKoN) *Rudolf Bannasch*, The German Bionics Competence Network "BioKoN" (supported by BioKoN) *Rolf Pfeifer*, Morphological computation: connecting body, brain and environment (supported by BioKoN) Robotics & Control: *Ruediger Dillmann*, Title to be announced *Vladimir V. Beletsky*, Regular and Chaotic Dynamics of Two-Legged Walking Robotics & Mechanics: *Martin Buehler*, Legged Robots Step Outside *Fumiya Iida*, Cheap Design Approach to Adaptive Behavior: Walking and Sensing through Body Dynamics With best regards, Hartmut Witte, Hiroshi Kimura, and Auke Ijspeert on behalf of the AMAM organizing committee. _ Call for last-minute-results posters:_ Researchers are invited to submit a one-page abstract describing their most recent results/developments. The call for abstracts is open to all (i.e. to people both with and without an accepted paper at the conference). There is a limit of one abstract/poster per participant as a first author. After a quick review by the conference chairs, authors of accepted abstracts will be invited to present their results as a poster in a special Last-Minute-Results session (short oral spotlight presentation + poster presentation). Note that the abstracts will not be included in the conference proceedings, but will be included in a special booklet. The* deadline for submission is September 10^th *, but can be submitted at any time from now on. We will try to review submissions and return a notification within one week. Please email your abstracts to amam at tu-ilmenau.de From oby at cs.tu-berlin.de Wed Aug 10 08:19:38 2005 From: oby at cs.tu-berlin.de (Klaus Obermayer) Date: Wed, 10 Aug 2005 14:19:38 +0200 (MEST) Subject: Connectionists: PhD and Postdoc positions Message-ID: Can you please post this job announcement on the list? Thanks and all the best Klaus =========================================================================== The Department for Electrical Engineering and Computer Science of the Technial University of Berlin (Germany) solicits applications for 1 Postdoctoral Position (BAT Ib) 1 Graduate Student Position (BAT IIa) (starting date October 1st 2005). The research tasks are in the field of machine learning / data mining and include the application and development of machine learning methods for developing adaptive and personalized e-commerce websites. Both positions are part of the joint project 'Cybersmart' in cooperation with several industrial and academic partners. Requirements: Candidates should hold a recent PhD-degree (postdoctoral position) or a Diplom or Master degree (graduate student position) in computer science, physics, mathematics, statistics, or related fields. Strong mathematical and programming skills, and research experience in the field of machine learning are required. The postdoctoral position will is also involved in the coordination of the project. The positions are initially for 2 years. Salary is according to the German BAT Ib (postdoctoral position) resp. BAT IIa (graduate student position) level. Application material (CV, list of publications, copies of certificates, summary of the doctoral resp. diploma thesis and two letters of reference) should be sent within two weeks after the release of this announcement to Prof. Dr. Klaus Obermayer, FR 2-1, Technische Universit?t Berlin, Franklinstra?e 28/29, 10587 Berlin, email: oby at cs.tu-berlin.de. For more information see: http://ni.cs.tu-berlin.de/. TUB seeks to increase the proportion of women and particularly encourages women to apply. Women will be preferred given equal qualifications. Disabled persons will be preferred given equal qualifications. From marc at memory.syr.edu Fri Aug 12 11:19:24 2005 From: marc at memory.syr.edu (Marc Howard) Date: Fri, 12 Aug 2005 11:19:24 -0400 (EDT) Subject: Connectionists: Tenure-track faculty position at Syracuse Univ Message-ID: Greetings Connectionists, Thought this opening might be of interest to some readers of this list. Thanks, marc -- marc at memory.syr.edu Marc Howard, Ph.D. Syracuse University Department of Psychology http://memory.syr.edu/ (315) 443-1864 ------------------------------------------------------------------------------ The Psychology Department of Syracuse University invites applications for a tenure-track position in Cognitive Psychology. Candidates with post-doctoral experience and with expertise in quantitative modeling or in the use of psychophysiological measures in any area of cognitive psychology (e.g., memory, learning, attention, emotion, perception, cognitive neuroscience) are encouraged to apply. Responsibilities of the position include 1) establishing or continuing to build a program of research in one of the areas mentioned; and 2) research supervision and teaching at the undergraduate and graduate level (e.g., courses in brain and cognition, quantitative methods, and/or in areas of research specialization). This position is at the assistant professor level although candidates at the advanced assistant level are encouraged to apply. Candidates with interest and expertise in teaching a graduate level course in statistics, and with substantive and statistical expertise that complement areas of strength in the department (social problems, cognitive aging, health, and child/school), are highly desirable. Collaborative research is a signature of the deparment and institution. Candidates who are interested in cross-area as well as interdisciplinary teaching and research are encouraged to apply. Successful candidates will evidence a commitment to involving students in an active research program, high quality scholarship and the potential for extramural funding. Additional information about the department may be found at http://psychweb.syr.edu. Submit a letter of application (including statements of teaching and research interests), a curriculum vita, three letters of recommendation, and evidence of teaching effectiveness to the Chair of the Cognitive Search Committee; 430 Huntington Hall, Department of Psychology, Syracuse University, Syracuse, NY 13244-2340. Applications received by November 1 will receive full consideration, with subsequent applications considered until the position is filled. Syracuse University is an equal opportunity, affirmative-action employer. The Psychology Department is committed to enhancing the diversity of its faculty and especially encourages applications from women, members of minority groups, and individuals with disabilities. From nicolas.lewy at unine.ch Mon Aug 15 11:03:04 2005 From: nicolas.lewy at unine.ch (LEWY Nicolas) Date: Mon, 15 Aug 2005 17:03:04 +0200 Subject: Connectionists: Paper on model of spoken word recognition in French Message-ID: <4300AEA8.6010804@unine.ch> The following article (written in French) on an interactive activation model of spoken word recognition in French has just been published on-line. Lwy, N., Grosjean, F., Grosjean, L., Racine, I. & Yersin, C. (2005). "Un modle psycholinguistique informatique de la reconnaissance des mots dans la chane parle du franais". Journal of French Language Studies, Volume 15, Issue 1, pages 2548. For (currently free) access via your WWW browser, use this URL: http://journals.cambridge.org/action/displayIssue?jid=JFL&volumeId=15&issueId=01 Abstract: Nous proposons un nouveau modle psycholinguistique informatique du franais. Le modle, intitul FN5, porte sur la reconnaissance de mots parls, prsents en isol (dterminant, adjectif antpos, substantif) ou en suites de deux mots (dterminant et substantif, adjectif antpos et substantif). Grce un lexique de plus de 17 000 mots, une architecture connexionniste localiste, et des mcanismes dvelopps spcifiquement (processeur de position, groupements de connexions, et point d'isolation), le modle peut simuler, de manire squentielle ou simultane, certains effets propres au mot isol (frquence, longueur, homophonie) ou une suite de mots (coupure lexicale, enchanement, liaison, effacement du schwa). De plus, le modle tient compte de certaines diffrences qui existent entre le franais standard et le franais de Suisse romande (nombre de voyelles, dure vocalique en fin de mot, et statut du schwa). Nous dcrivons ce modle, prsentons son interface graphique, et l'illustrons avec des exemples de simulation. Thank you for your interest / Merci de votre intrt, Nicolas Lwy Laboratoire de traitement du langage et de la parole (Language and Speech Processing Laboratory) Universit de Neuchtel, Avenue du Premier-Mars 26 CH-2000 Neuchtel, Switzerland email: nicolas.lewy at unine.ch From jcortes at onsager.ugr.es Tue Aug 16 03:04:16 2005 From: jcortes at onsager.ugr.es (J.M. Cortes) Date: Tue, 16 Aug 2005 09:04:16 +0200 (CEST) Subject: Connectionists: paper available on fast presynaptic noise on ANN Message-ID: Dear all, I would like to announce the following paper, that can be downloaded from arxiv.org/pdf/q-bio.NC/0508013 and that will be published in Neural Computation this year 2005. Effects of fast presynaptic noise in attractor neural networks J. M. Cortes, J. J. Torres, J. Marro, P. L. Garrido, H. J. Kappen We study both analytically and numerically the effect of presynaptic noise on the transmission of information in attractor neural networks. The noise occurs on a very short-time scale compared to that for the neuron dynamics and it produces short-time synaptic depression. This is inspired in recent neurobiological findings that show that synaptic strength may either increase or decrease on a short-time scale depending on presynaptic activity. We thus describe a mechanism by which fast presynaptic noise enhances the neural network sensitivity to an external stimulus. The reason for this is that, in general, the presynaptic noise induces nonequilibrium behavior and, consequently, the space of fixed points is qualitatively modified in such a way that the system can easily scape from the attractor. As a result, the model shows, in addition to pattern recognition, class identification and categorization, which may be relevant to the understanding of some of the brain complex tasks. Jesus M. Cortes, PostDoctoral Researcher SNN, Radboud University Nijmegen The Netherlands Tel. +31 243614230 Fax. +31 243541435 From nicolas.lewy at unine.ch Mon Aug 15 14:25:22 2005 From: nicolas.lewy at unine.ch (LEWY Nicolas) Date: Mon, 15 Aug 2005 20:25:22 +0200 Subject: Connectionists: Paper on model of spoken word recognition in French (* corrected *) Message-ID: <4300DE12.5040701@unine.ch> (* The previous version of this message has unfortunately been stripped of accented characters. Please read this corrected message instead *) The following article (written in French) on an interactive activation model of spoken word recognition in French has just been published on-line. Lewy, N., Grosjean, F., Grosjean, L., Racine, I. & Yersin, C. (2005). "Un modele psycholinguistique informatique de la reconnaissance des mots dans la chaine parlee du francais". Journal of French Language Studies, Volume 15, Issue 1, pages 2548. For (currently free) access via your WWW browser, use this URL: http://journals.cambridge.org/action/displayIssue?jid=JFL&volumeId=15&issueId=01 Abstract: Nous proposons un nouveau modele psycholinguistique informatique du francais. Le modele, intitule' FN5, porte sur la reconnaissance de mots parles, presentes en isole (determinant, adjectif antepose, substantif) ou en suites de deux mots (determinant et substantif, adjectif antepose et substantif). Grace a un lexique de plus de 17 000 mots, a une architecture connexionniste localiste, et a des mecanismes developpes specifiquement (processeur de position, groupements de connexions, et point d'isolation), le modele peut simuler, de maniere sequentielle ou simultanee, certains effets propres au mot isole (frequence, longueur, homophonie) ou a une suite de mots (coupure lexicale, enchainement, liaison, effacement du schwa). De plus, le modele tient compte de certaines differences qui existent entre le francais standard et le francais de Suisse romande (nombre de voyelles, duree vocalique en fin de mot, et statut du schwa). Nous decrivons ce modele, presentons son interface graphique, et l'illustrons avec des exemples de simulation. Thank you for your interest / Merci de votre interet, Nicolas Lewy Laboratoire de traitement du langage et de la parole (Language and Speech Processing Laboratory) Universite de Neuchatel, Avenue du Premier-Mars 26 CH-2000 Neuchatel, Switzerland email: nicolas.lewy at unine.ch From oza at email.arc.nasa.gov Mon Aug 15 19:05:26 2005 From: oza at email.arc.nasa.gov (Nikunj Oza) Date: Mon, 15 Aug 2005 16:05:26 -0700 Subject: Connectionists: CFP: Information Fusion Journal - Special Issue on Applications of Ensemble Methods Message-ID: <43011FB6.6050809@email.arc.nasa.gov> APOLOGIES FOR MULTIPLE COPIES Call for papers for a special issue of Information Fusion An International Journal on Multi-Sensor, Multi-Source Information Fusion An Elsevier Publication On APPLICATIONS OF ENSEMBLE METHODS Editor-in-Chief: Dr. Belur V. Dasarathy, FIEEE d.belur at elsevier.com http://belur.no-ip.com Guest Editors: Nikunj C. Oza, Kagan Tumer The Information Fusion Journal is planning a special issue devoted to Applications of Ensemble Methods in Machine Learning and Pattern Recognition. Ensembles, also known as Multiple Classifier Systems (MCSs) and Committee Classifiers, were originally motivated by the desire to avoid relying on just one learned model when only a small amount of training data is available. Because of this, most studies on ensembles have evaluated their new algorithms on relatively small datasets; most notably, datasets from the University of California, Irvine (UCI) Machine Learning Repository. However, modern data mining problems raise a variety of issues very different from the ones ensembles have traditionally addressed. These new problems include too much data; data that are distributed, are noisy, and represent changing environments; and performance measures different from the standard accuracy measurements; among others. The aim of this issue is to examine the different applications that raise these modern data mining problems, and how current and novel ensemble methods aid in solving these problems. Manuscripts (which should be original and not previously published or presented even in a more or less similar form under any other forum) covering new applications as well as the theories and algorithms of ensemble learning algorithms developed to address these applications are invited. Contributions should be described in sufficient detail to be reproducible on the basis of the material presented in the paper. Topics appropriate for this special issue include, but are not limited to: ? Innovative applications of ensemble methods. ? Novel algorithms that address unique requirements (for example, different performance measures or running time constraints) of an application or a class of applications. ? Novel theories developed under assumptions unique to an application or a class of applications. ? Novel approaches to distributed model fusion. Manuscripts should be submitted electronically online at http://ees.elsevier.com/inffus (The corresponding author will have to create a user profile if one has not been established before at Elsevier.) Please also send without fail an electronic copy to oza at email.arc.nasa.gov (PDF format preferred), Guest Editors Nikunj C. Oza and Kagan Tumer NASA Ames Research Center Mail Stop 269-3 Moffett Field, CA 94035-1000, USA Deadline for Submission: November 30, 2005 -- -------------------------------------- Nikunj C. Oza, Ph.D. Tel: (650)604-2978 Research Scientist Fax: (650)604-4036 NASA Ames Research Center E-mail: oza at email.arc.nasa.gov Mail Stop 269-3 Web: http://ic.arc.nasa.gov/people/oza Moffett Field, CA 94035-1000 USA From wduch at phys.uni.torun.pl Tue Aug 16 08:28:41 2005 From: wduch at phys.uni.torun.pl (wduch@phys.uni.torun.pl) Date: Tue, 16 Aug 2005 14:28:41 +0200 Subject: Connectionists: Workshop: Creating an Artificial Brain, Sept 15, Torun, Poland Message-ID: <1124195321.4301dbf9339b2@poczta.umk.pl> ‘Creating an Artificial Brain’ Organisers: W Duch & JG Taylor Date: Sep 15, 2005, after ICANN 2005 Place: Nicolaus Copernicus University, Torun, Poland There is an upsurge of interest in the attempt to build an artificial brain. This is for several reasons: i) The enormous increase of computing power over the last decade or so; ii) The increased understanding of the brain arising from developments in brain imaging and single cell recordings and the associated use of probing psychological paradigms; iii) The increased demands from industry in terms of the creation of autonomous agents with some form of cognitive powers, even up to that of ‘conscious’ machines. An important aspect of this general advance is that guidance is being taken from the higher processing power of the brain, to help produce systems with a similar or even greater power. In so doing an overlap is now growing between machine intelligence per se and computational neuroscience, with projects being guided by global brain processing as observed by neuroscience. Already several such projects, using clusters of up to ten thousand nodes are being pursued; with the largest such cluster of 150,000 nodes indicating what is now available. This workshop proposes to explore the issues raised in following such an avenue: 1) The nature of the global brain itself, and the data relevant to help guide us forward; 2) The problems of the level at which to base any such approach, in terms of neuron complexity, connectivity levels, number of separate modules, learning rules, use of neuro-modulators, range of brain science data to be explained or incorporated in any model; 3) The problems of size and speed of the resulting computation, and whether to aim for real-time or off-line learning and responses 4) How to incorporate the faculties of attention, emotion and memory in an input-output system able to develop concepts and to ultimately be able to ‘think for itself’. Applications are welcome for proposed talks in the workshop (which will last one day). Please send suggestions (title, author co-ordinates and abstract) to the organizers at john.g.taylor at kcl.ac.uk and ASWDuch at ntu.edu.sg. We look forward to seeing you at the Workshop! For details on the venue and organization please see: http://www.ibspan.waw.pl/ICANN-2005/workshops.html From cindy at bu.edu Wed Aug 17 12:18:04 2005 From: cindy at bu.edu (Cynthia Bradford) Date: Wed, 17 Aug 2005 12:18:04 -0400 Subject: Connectionists: Neural Networks 18(7) 2005 Message-ID: <200508171618.j7HGI4UN002931@kenmore.bu.edu> NEURAL NETWORKS 18(7) Contents - Volume 18, Number 7 - 2005 ------------------------------------------------------------------ *****Psychology and Cognitive Science***** Elman topology with sigma-pi units: An application to the modeling of verbal hallucinations in schizophrenia Juan C. Valle-Lisboa, Florencia Reali, Hector Anastasia, and Eduardo Mizraji Linear recursive distributed representations Thomas Voegtlin and Peter F. Dominey *****Mathematical and Computational Analysis***** Generalized hamming networks and applications Konstantinos Koutroumbas and Nicholas Kalouptsidis Simultaneous L-p approximation order for neural networks Zong-Ben Xu and Fei-Long Cao Stochastic complexities of reduced rank regression in Bayesian estimation Miki Aoyagi and Sumio Watanabe *****Engineering and Design***** Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms Konstantinos P. Ferentinos Sensitivity analysis applied to the construction of radial basis function networks D. Shi, D.S. Yeung, and J. Gao A robust classifier combined with an auto-associative network for completing partly occluded images Takashi Takahashi and Takio Kurita Data-partitioning using the Hilbert space filling curves: Effect on the speed of convergence of Fuzzy ARTMAP for large database problems Jose Castro, Michael Georgiopoulos, Ronald Demara, and Avelino Gonzalez *****Technology and Applications***** Wavelet neural network classification of EEG signals by using AR model with MLE preprocessing Abdulhamit Subasi, Ahmet Alkan, Etem Koklukaya, and M. Lemal Kiymik Using ICA for removal of ocular artifacts in EEG recorded from blind subject Arthur Flexer, Herbert Bauer, Jurgen Pripfl, and Georg Dorffner *****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 Y 11,000 (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- membership without $30 SEK 200 not available Neural Networks to non-student (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 2810 Crossroads Drive, Suite 3800 Madison WI 53718 USA 608 443 2461, ext. 138 (phone) 608 443 2474 (fax) srees at reesgroupinc.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 JNNS Secretariat c/o Fuzzy Logic Systems Institute 680-41 Kawazu, Iizuka Fukuoka 820-0067 Japan 81 948 24 2771 (phone) 81 948 24 3002 (fax) jnns at flsi.cird.or.jp http://www.jnns.org/ ---------------------------------------------------------------------------- From M at meeter.nl Thu Aug 18 07:51:32 2005 From: M at meeter.nl (M@meeter.nl) Date: Thu, 18 Aug 2005 13:51:32 +0200 (CEST) Subject: Connectionists: papers on models of the hippocampus Message-ID: <3114.130.37.96.138.1124365892.squirrel@webmail.internl.net> Connectionists may be interested in the following four papers on the hippocampus and memory. Of all papers final drafts are available at www.cs.vu.nl/~cogsci/cogpsy/meeter. * Meeter, M., Talamini, L.M. & Murre, J.M.J. (2004). Mode shifting between storage and recall based on novelty detection in oscillating hippocampal circuits. Hippocampus, 14, 722-741. Presents a relatively low-level model of the hippocampus. Building on ideas of Hasselmo, effects of novelty and Acetylcholine (ACh) are explored. In particular, the model shows that the relatively slow time course of ACh's effects in the hippocampus are consistent with the time scale of learning. http://www3.interscience.wiley.com/cgi-bin/fulltext/107633281/PDFSTART * Meeter, M. Talamini, L.M., Schmitt, J.A.J., & Riedel, W.J. (in press). Effects of 5-HT on memory and the hippocampus: model and data. Neuropsychopharmacology. The model of Meeter et al. (2004) applied to serotonin's effects on memory. http://www.cs.vu.nl/~cogsci/cogpsy/meeter/p_publications_files/meeteretal-npp-web-jul05.pdf * Talamini, L.M., Meeter, M., Murre, J.M.J., Elvevag, B. & Goldberg, T.E. (2005). Reduced parahippocampal connectivity produces schizophrenia-like deficits in simulated neural circuits with reduced parahippocampal connectivity. Archives of General Psychiatry, 62, 485-493. A more abstract model of the medial temporal lobe and episodic memory is presented. The model suggests that abnormalities in entorhinal connectivity may cause the episodic memory deficits seen in schizophrenia (the rather silly title was a last-minute "improvement" by the journal). http://archpsyc.ama-assn.org/cgi/reprint/62/5/485 * Meeter, M., Myers, C.E., & Gluck, M.A. (2005). Integrating incremental learning and episodic memory models of the hippocampal region. Psychological Review, 112, 560-585. Theories of episodic memory and of incremental learning make mutually exclusive claims on the hippocampus. We present a model of the medial temporal lobe that reconciles these claims, and incorporates a new theory of familiarity. http://www.cs.vu.nl/~cogsci/cogpsy/meeter/p_publications_files/meeteretal-psychrev-web-dec04.pdf Best, Martijn Meeter ------------------------------ Dr. M. Meeter Dept. of Cognitive Psychology Vrije Universiteit Amsterdam The Netherlands ------------------------------ From pli at richmond.edu Thu Aug 18 15:30:23 2005 From: pli at richmond.edu (Ping Li) Date: Thu, 18 Aug 2005 15:30:23 -0400 (EDT) Subject: Connectionists: papers available Message-ID: <49809.141.166.178.153.1124393423.squirrel@141.166.178.153> Dear Colleague, You might be interested in the following three papers: Li, P., Farkas, I., & MacWhinney (2004). Early lexical acquisition in a self-organizing neural network. Neural Networks, 17, 1345-1362. Hernandez, A., Li, P., & MacWhinney, B. (2005). The emergence of competing modules in bilingualism. Trends in Cognitive Sciences, 9, 220-225. Li, P., & Cunningham, K. (2005). The APA Style Converter: A Web-based interface for converting articles to APA Style for publication. Behavior Research Methods, 37, 219-223. The papers may be requested for download at http://cogsci.richmond.edu/publications.html Sincerely Ping Li From rodrigo at neuron.ffclrp.usp.br Thu Aug 18 16:23:00 2005 From: rodrigo at neuron.ffclrp.usp.br (Rodrigo Freire Oliveira) Date: Thu, 18 Aug 2005 17:23:00 -0300 Subject: Connectionists: Second Call: LASCON*2006 Message-ID: <4304EE24.6020006@neuron.ffclrp.usp.br> Second Call - LASCON*2006 the I Latin American School on Computational Neuroscience LASCON 2006 http://neuron.ffclrp.usp.br/lascon/home.htm January 15-28 2006 University of Sao Paulo Ribeirao Preto, SP Brazil Faculty David Beeman, University of Colorado, Boulder, CO, USA James Bower, University of Texas, San Antonio, TX, USA Kim Blackwell, George Mason University, Fairfax, VA, USA Michael Hines, Yale University, New Haven, CT, USA Michael Hasselmo, Boston University, Boston, MA, USA Dieter Jaeger, Emory University, Atlanta, GA, USA Roland Koberle, University of Sao Paulo, Sao Carlos, SP, Brazil Marcelo Mazza, University of Sao Paulo, Ribeirao Preto, SP, Brazil Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Reynaldo Pinto, University of Sao Paulo, Sao Paulo, SP, Brazil Arnd Roth, University College, London, UK Michael Vanier, CALTECH, Pasadena, CA, USA Charles Wilson, University of Texas, San Antonio, TX, USA Organizer Antonio Roque, University of Sao Paulo, Ribeirao Preto, SP, Brazil Co-organizer Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Scientific Committee David Beeman, University of Colorado, Boulder, CO, USA James Bower, University of Texas, San Antonio, TX, USA Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Deadline for student application: September 9, 2005 (Friday) The I Latin American School on Computational Neuroscience (LASCON 2006) aims at introducing advanced undergraduate and graduate students to the use of methods for detailed modeling of neurons and neural circuits, based on the Hodgkin-Huxley formalism, the cable equation and the compartmental modeling technique. The use of these methods will be illustrated with the development and investigation of numerical simulations with the programs GENESIS and NEURON. The school is divided in two weeks, the first one for theoretical lectures and hands on tutorials, and the second one for invited lectures and the development of project works by the students. Students will follow a highly demanding schedule of morning lectures followed by afternoon and evening computational laboratory sessions. The School will be held at the campus of the University of S?o Paulo at Ribeir?o Preto (a city about 313 km north of Sao Paulo). Course applicants should be fluent in English (written/spoken) and have a solid background in life and/or hard sciences (some experience in computer programming is also desirable). Applications are welcome and should be made by using the application form on the school web page (http://neuron.ffclrp.usp.br/lascon/home.htm). The number of students is limited to 20 and priority will be given to students from Latin American countries. Under exceptional circumstances, students from other areas of the world could be accepted as well. Costs for accommodation and meals will be covered by the school organization. In exceptional cases, limited funding is available to partly cover travel expenses. More information on the school can be found at http://neuron.ffclrp.usp.br/lascon/home.htm From mo at ecs.soton.ac.uk Mon Aug 22 12:30:03 2005 From: mo at ecs.soton.ac.uk (Manfred Opper) Date: Mon, 22 Aug 2005 17:30:03 +0100 Subject: Connectionists: postdoctoral position in Southampton (ERRATUM) Message-ID: <4309FD8B.7306300B@ecs.soton.ac.uk> In our job posting for the project "Inference in Complex Stochastic Dynamic Environmental Models" at the ISIS group in Southampton we gave the WRONG link http://www.jobs.soton.ac.uk/adminweb/jsp/jobs/sJobview.jsp?function=View&id=04R0842 for the application details. The CORRECT link is http://www.jobs.soton.ac.uk/adminweb/jsp/jobs/sJobview.jsp?function=View&id=04R0854 Could any person who has applied with the WRONG REF NUMBER 04R0842 please send me an email (mo at ecs.soton.ac.uk), so that we can redirect their applications? Sorry for the mistake Manfred Opper From gbrown at cs.man.ac.uk Wed Aug 24 05:15:08 2005 From: gbrown at cs.man.ac.uk (Gavin Brown) Date: Wed, 24 Aug 2005 10:15:08 +0100 Subject: Connectionists: PhD studentship at Manchester Message-ID: <430C3A9C.7030500@cs.man.ac.uk> ======================================================================= A three-year, fully-funded PhD studentship is available in the AI group at the University of Manchester, UK. The following project is suggested, though additional studentships are available in the general area of Machine Learning. Application details below. ============= "Interactions between Learning Classifier Systems and Ensemble Systems" For best generalisation ability, an ensemble of predictors must exhibit a "diversity" of errors---for majority voting ensembles, how to define this concept remains an open research question. The problem can be seen as one of credit assignment, judging how much a particular predictor contributes to the overall classification error. In spite of the shortage of theoretical results, ensembles typically show significant performance improvements over single predictors. Learning Classifier Systems (LCS) use evolutionary computation techniques to generate groups of rules---the decisions of these rules are combined by a weighted voting mechanism. Recent works in the literature have extensively studied credit assignment in LCS, providing significant insights---in spite of this and other contributions, LCS remain underused in the wider ML community. The aim of the project is to transfer theoretical results and empirical techniques between these two fields, providing both with new insights into how to understand their learning systems, and new mechanisms for solving their datasets. A priority is to replace current heuristics with theoretically justified mechanisms, so there will be a significant (but not exclusive) emphasis on the mathematical elements of the project. ============= Applicants should hold a first or upper second class degree in computer science, mathematics, physics, or similarly numerate disciplines. Additional research experience such as during a Master's degree will be seen as a significant advantage. A solid background in programming (e.g. Java/C++) is also essential. For further details and informal discussions, email Dr Gavin Brown: Email: gavin.brown at manchester.ac.uk Web : http://www.cs.man.ac.uk/~gbrown/ Those interested in the position should send detailed CVs, including: 1) full academic history, 2) description of motivations and research interests, 3) contact details for two referees, either by email (preferable) or to the following address: Dr Gavin Brown University of Manchester School of Computer Science, Kilburn Building, Manchester, M13 9LP ======================================================================= From terry at salk.edu Thu Aug 25 14:27:26 2005 From: terry at salk.edu (Terry Sejnowski) Date: Thu, 25 Aug 2005 11:27:26 -0700 Subject: Connectionists: NEURAL COMPUTATION 17:10 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 17, Number 10 - October 1, 2005 NOTE Gradient-Based Adaptation of General Gaussian Kernels Tobias Glasmachers and Christian Igel LETTERS Learning Only When Necessary: Better Memories of Correlated Patterns in Networks with Bounded Synapses Walter Senn and Stefano Fusi Coding of Temporally Varying Signals in Networks of Spiking Neurons with Global Delayed Feedback Naoki Masuda, Brent Doiron, Andre Longtin and Kazuyuki Aihara Attention-Gated Reinforcement Learning of Internal Representations for Classification Pieter R. Roelfsema and Arjen van Ooyen Computing with Continuous Attractors: Stability and Online Aspects Si Wu and Shun-ichi Amari Optimal Signal Estimation in Neuronal Models Petr Lansky and Priscilla E. Greenwood Finite State Automata Resulting from Temporal Information Maximization and a Temporal Learning Rule Thomas Wennekers and Nihat Ay Fast Access to Concepts in Concept Lattices via Bidirectional Associative Memories Rohana Rajapakse and Michael Denham ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2005 - VOLUME 17 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $114 $54 $57.78 Individual $100 $107.00 $143 $90 $96.30 Institution $680 $727.60 $734 $612 $654.84 * 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 Wael.El-deredy at manchester.ac.uk Thu Aug 25 07:33:23 2005 From: Wael.El-deredy at manchester.ac.uk (Wael El-Deredy) Date: Thu, 25 Aug 2005 12:33:23 +0100 Subject: Connectionists: PhD studentship in graphical models Message-ID: <20050825123323281.00000001288@8VJ9J1J> Bayesian networks for knowledge discovery and collaborative filtering EPSRC CASE Studentship (see new eligibility criteria below) A PhD studentship is available for three years to develop and evaluate graphical models for collaborative filtering. Probabilistic recommender systems capable of hidden variables underpinning preference, choice and purchase behaviour will be developed within the Bayesian belief networks framework. The project will involve close interaction with industrial sponsors and will be supervised by Professor Paulo Lisboa at the school of Computing and Mathematical Sciences - Liverpool John Moores University and Dr. Wael El-Deredy at School of Psychological Sciences, University of Manchester. The successful candidates will receive an enhanced EPSRC stipend. Eligible candidates* wishing to apply should send by email a supporting statement together with a CV containing names and addresses of two referees by the closing date of Monday 26th September 2005. * Candidates must satisfy the EPSRC eligibility requirements: Recent changes to the criteria have opened full CASE studentship funding to EU nationals with residence in the UK for at least 3 years, which now include periods of stay for the purpose of higher education. For further information and applying contact Professor PJG Lisboa email: p.j.lisboa at livjm.ac.uk, tel. 0151 231 2225 --------- Wael.El-Deredy at Manchester.ac.uk School of Psychological Sciences University of Manchester Zochonis Building - Oxford Road Manchester M13 9PL UK Tel. +44(0) 161 275 2566 Fax +44(0) 161 275 2685 From hdaume at ISI.EDU Tue Aug 23 14:19:01 2005 From: hdaume at ISI.EDU (Hal Daume III) Date: Tue, 23 Aug 2005 11:19:01 -0700 (PDT) Subject: Connectionists: CFP: Bayesian Methods in NLP workshop at NIPS Message-ID: ************************************************************************ CALL FOR PAPERS Bayesian Methods for Natural Language Processing Workshop at the Neural Information Processing Systems Conference (NIPS 2005) http://www.isi.edu/~hdaume/BayesNLP/ ** Submission Deadline: 21 October 2005 ** ************************************************************************ [ Apologies for multiple postings ] OVERVIEW -------- Models of natural language processing problems are often incredibly complex, and there is never enough data to properly estimate all the required parameters. This has lead to a strong need for learning techniques with built-in capacity control; most classical solutions to this problem involve largely ad-hoc smoothing techniques. The application of Bayesian learning methods to these problems could potentially result in more effective models, for which extensive cross-validation is no longer required for hyperparameter tuning or model selection. The goals of this workshop are to bring together researchers from both the Bayesian machine learning community and the natural language processing community to enable cross-fertilization of techniques, models and applications. We wish to focus on the following issues: * Statistical Models: Current Bayesian models for text have largely focused on "bag of words" style approaches, where conditional independence is assumed between words. This leads to a convenient interpretation of a document as a sequence of draws from multinomial distributions, but does not account for any of the internal structure that exists in documents and which NLP researchers are interested in. How can we build models that move beyond the bag of words assumption? What structures are useful for modeling? How can we model these structures efficiently? Can we learn these models automatically? * Applications-oriented Models: Many statistical models for text have aimed at automatically inferring implicit relationship between varied elements of documents in a corpus. How can we use such models to aid in applications? Can we develop similar models that are aimed at solving a real-world NLP task? For what NLP applications are Bayesian techniques appropriate and how can we develop models specific to these problems? CALL FOR PARTICIPATION ---------------------- We invite submission of workshop papers that discuss ongoing or completed work dealing with Bayesian techniques applied to natural language processing problems (see below for an incomplete list of possible topics). A workshop paper should be no more than six pages in the standard NIPS format. Authorship should not be blind. Please submit a paper by emailing it in Postscript or PDF format to hdaume at isi.edu with the subject line "BNLP Submission". We anticipate accepting four to six such papers for 15 minute presentation slots (exact details will be worked out shortly). Please only submit an article if at least one of the authors will be able to attend the workshop and present the work. We are especially interested in submissions from authors in the NLP community who have not previously attended a NIPS conference. If you fall into this category, please note this in your email when you submit your paper. Relevant Topics: * Models that move beyond the bag-of-words assumption * Techniques that apply to problems other than language modeling * Structure-learning techniques for language * Bayesian extensions to well-known NLP models * Application of Bayesian techniques to NLP problems * Both supervised and unsupervised techniques are welcome We also welcome position papers of at most two pages in length that discuss, with appropriate argumentation, whether or not Bayesian techniques are applicable to NLP problems and, if so, which ones. These should be submitted in the same way as standard workshop papers. These will be used to help guide discussion during panel sessions. IMPORTANT DATES --------------- 18 Aug 05 -- Call for participation 21 Oct 05 -- Paper submission deadline 4 Nov 05 -- Notification of paper acceptance 25 Nov 05 -- Survey and position paper deadlines 9/10 Dec 05 -- Workshop in Whistler RESEARCHER SURVEY ----------------- Regardless of whether you submit a paper or not, if you are a researcher in either the Bayesian learning community or the NLP community, please complete our survey (available on the web page), which will serve to guide the panel discussions at the workshop. ORGANIZATION ------------ Hal Daume III Information Sciences Institute hdaume at isi.edu http://www.isi.edu/~hdaume/ Yee Whye Teh National University of Singapore tehyw at comp.nus.edu.sg http://www.cs.berkeley.edu/~ywteh/ From M.Denham at plymouth.ac.uk Wed Aug 31 06:50:34 2005 From: M.Denham at plymouth.ac.uk (Mike Denham) Date: Wed, 31 Aug 2005 11:50:34 +0100 Subject: Connectionists: PhD position (Research Assistant) in Computational Neuroscience Message-ID: <52A8091888A23F47A013223014B6E9FE060A8504@03-CSEXCH.uopnet.plymouth.ac.uk> University of Plymouth Faculty of Science Centre for Theoretical and Computational Neuroscience Research Assistant - Three Year Fixed Term Appointment (salary range ?14,751 - ?17,601 pa) Applications are invited for a 3-year PhD position (Research Assistant) in computational neuroscience in the Centre for Theoretical and Computational Neuroscience at the University of Plymouth, UK. The post is funded as part of an EU-funded Integrated Project: "FACETS: Fast Analog Computing with Transient States in Neural Architectures". The stated objective of FACETS is to explore and exploit the yet unknown computational principles that constitute the basis of information processing in the brain. The project involves experimental neuroscience, the construction of models and analytical descriptions for neural cells and networks and the construction of very large scale neural circuits in VLSI technology. The FACETS consortium includes fifteen of the major laboratories in Europe in these areas. The PhD student in Plymouth will work specifically on the construction of computational models of neural processing of visual information in the neocortex, in collaboration with the other participants in the project. Maintaining a close level of collaboration will require short visits to the laboratories of the project collaborators. The normal UK/EU PhD tuition fees will be funded for the duration of the three-year appointment, together with a salary in the range ?14,751 - ?17,601 per annum. Applicants for the post must have a sound educational background at first degree level in a relevant subject area, with a high standard of achievement consistent with entering into a PhD programme. They should also possess a computing/mathematics background appropriate to carrying out research in computational modelling of the brain. Informal enquiries should be made to Professor Mike Denham, Centre for Theoretical and Computational Neuroscience, University of Plymouth, Drake Circus, Plymouth, PL4 8AA, UK; tel: +44 (0)1752 232547; email: mdenham at plym.ac.uk From pli at richmond.edu Wed Aug 31 11:02:59 2005 From: pli at richmond.edu (Ping Li) Date: Wed, 31 Aug 2005 11:02:59 -0400 Subject: Connectionists: cognitive neuroscience position Message-ID: <520A2447-1A30-11DA-A4AD-0011246FEB0E@richmond.edu> The University of Richmond seeks to fill a tenure track position in Neuroscience at the assistant professor level in the Department of Psychology. Although we are particularly interested in individuals with a specialization in human cognitive or behavioral neuroscience, we will consider outstanding applicants in all areas. Faculty are expected to maintain a vigorous research program, actively engage undergraduates in substantive research, attract extramural support, and commit to highly effective teaching at all levels of the undergraduate curriculum. Candidates should have completed the Ph.D. degree by the August 2006 starting date. Send vita, statements of research plan and teaching philosophy, copies of transcripts, and three letters of reference to Scott Allison, Search Coordinator, Department of Psychology, University of Richmond, VA 23173. Review of applications will commence Oct. 1, 2005 and continue until the position is filled. The University of Richmond is a highly selective private university with approximately 3500 students located on a beautiful campus six miles west of the heart of Richmond and in close proximity to the ocean, mountains, and Washington, D.C. The University of Richmond values diversity in its faculty, staff, and student body. In keeping with this commitment, our academic community strongly encourages applications from diverse candidates and candidates who support diversity. For more information please go to http://psychology.richmond.edu. ---------------------------------------------------------------------- Ping Li, Ph.D. Associate Professor Graduate Program Coordinator Department of Psychology University of Richmond Richmond, VA 23173, USA Email: pli at richmond.edu http://www.richmond.edu/~pli/ http://cogsci.richmond.edu/ Bilingualism: Language and Cognition: http://cogsci.richmond.edu/bilingualism/bilingualism.html ----------------------------------------------------------------------- From mark at paskin.org Wed Aug 31 13:42:20 2005 From: mark at paskin.org (Mark A. Paskin) Date: Wed, 31 Aug 2005 10:42:20 -0700 Subject: Connectionists: CFP: NIPS 2005 Workshop: Intelligence Beyond the Desktop Message-ID: ################################################################ CALL FOR PARTICIPATION Intelligence Beyond the Desktop a workshop at the 2005 Neural Information Processing Systems (NIPS) Conference Submission deadline: Friday, October 14, 2005 http://ai.stanford.edu/~paskin/ibd05/ ################################################################ OVERVIEW We are now well past the era of the desktop computer. Trends towards miniaturization, wireless communication, and increased sensing and control capabilities have led to a variety of systems that distribute computation, sensing, and controls across multiple devices. Examples include wireless sensor networks, multi-robot systems, networks of smartphones, and large area networks. Machine learning problems in these non-traditional settings cannot faithfully be viewed in terms of a data set and an objective function to optimize; physical aspects of the system impose challenging new constraints. Resources for computation and actuation may be limited and distributed across many nodes, requiring significant coordination; limited communication resources can make this coordination expensive. The scale and complexity of these systems often leads to large amounts of structured data that make state estimation challenging. In addition, these systems often have other constraints, such as limited power, or under-actuation, requiring reasoning about the system itself during learning and control. Furthermore, large-scale distributed systems are often unreliable, requiring algorithms that are robust to failures and lossy communication. New learning, inference, and control algorithms that address these challenges are required. This workshop aims to bring together researchers to discuss new applications of machine learning in these systems, the challenges that arise, and emerging solutions. FORMAT This one-day workshop will consist of invited talks and talks based upon submitted abstracts, with some time set aside for discussion. Our (tentative) invited speakers are: * Dieter Fox (University of Washington) * Leonidas Guibas (Stanford University) * Sebastian Thrun, (Stanford University) will speak about the machine learning algorithms used in Stanley, Stanford's entry into the DARPA Grand Challenge. CALL FOR PARTICIPATION Researchers working at the interface between machine learning and non- traditional computer architectures are invited to submit descriptions of their research for presentation at the workshop. Of particular relevance is research on the following topics: * distributed sensing, computation, and/or control * coordination * robustness * learning/inference/control under resource constraints (power, computation, time, etc.) * introspective machine learning (reasoning about the system architecture in the context of learning/inference/control) We especially encourage submissions that address unique challenges posed by non-traditional architectures for computation, such as * wireless sensor networks * multi-robot systems * large-area networks Submissions should be extended abstracts in PDF format which are no longer than three (3) pages long in 10pt or larger font. Submissions may be e-mailed to ibd-2005 at cs.cmu.edu with the subject "IBD SUBMISSION". We plan to accept four to six submissions for 25 minute presentation slots. In your submission please indicate if you would present a poster of your work (in case there are more qualified submissions than speaking slots). Call for participation: Wednesday, August 31, 2005 Submission deadline: Friday, October 14, 2005 11:59 PM PST Acceptance notification: Tuesday, November 1, 2005 Workshop: Friday, December 9, 2005 Organizers * Carlos Guestrin (http://www.cs.cmu.edu/~guestrin/) * Mark Paskin (http://paskin.org) Please direct any inquiries regarding the workshop to ibd-2005 at cs.cmu.edu. From steve at cns.bu.edu Wed Aug 31 20:29:49 2005 From: steve at cns.bu.edu (Stephen Grossberg) Date: Wed, 31 Aug 2005 20:29:49 -0400 Subject: Connectionists: watercolor illusion and neon color spreading Message-ID: The following article is now available at http://www.cns.bu.edu/Profiles/Grossberg Baingio Pinna, and Stephen Grossberg (2005). The watercolor illusion and neon color spreading: A unified analysis of new cases and neural mechanisms. Journal of the Optical Society of America A, in press. ABSTRACT Coloration and figural properties of neon color spreading and the watercolor illusion are studied using phenomenal and psychophysical observations. Coloration properties of both effects can be reduced to a common limiting condition, a nearby color transition called the "two-dot limiting case", that clarifies their perceptual similarities and dissimilarities. The results are explained by the FACADE neural model of biological vision. The model proposes how local properties of color transitions activate spatial competition among nearby perceptual boundaries, with boundaries of lower contrast edges weakened by competition more than boundaries of higher contrast edges. This asymmetry induces spreading of more color across these boundaries than conversely. The model also predicts how depth and figure-ground effects are generated in these illusions. From silvia at sa.infn.it Mon Aug 1 08:56:43 2005 From: silvia at sa.infn.it (Silvia Scarpetta) Date: Mon, 01 Aug 2005 14:56:43 +0200 Subject: Connectionists: opening position in neural computation in Salerno Message-ID: OPENING A two-year postdoctoral position (Assegno di Ricerca) in Neural Networks / NeuroPhysics starting on next Fall 2005 (around October, 2005) is available at the Dept. of Physics "E.R. Caianiello" of the Universit degli Studi di Salerno, Italy, in the group headed by Prof. Maria Marinaro (http://www.sa.infn.it/NeuralGroup/) The monthly net amount of the fellowship is about 1200 Euro. RESEARCH ACTIVITY Areas of particular interest include: 1) Neurobiological applications of theoretical physics tools, computational and mathematical modeling of neural dynamics, cortical dynamics and oscillations, learning memory and modelling of STDP, rhythmic locomotion and related topics. 2) Neural networks applied to signal and image processing and to speech recognition. Description of current project in our group can be found in: http://www.sa.infn.it/NeuralGroup/ MINIMUM REQUIREMENTS master degree in scientific disciplines (physics, neuroscience, mathematics, computer science, engeneering, etc) and a PhD degree or a 3 years research experience in fields related with the position topic. Sufficient knowledge of the English language and of a programming language (C or Matlab) will be required for successful project work. *** Provide, please, a CV with publication list and description of research interests, and arrange for at least two letters of reference to be sent to: Prof. Maria Marinaro Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende I-84081 Baronissi (SA) Italy ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Prof. Maria Marinaro Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende 84081 Baronissi (SA) Italy Tel. +39 089 965318 Web page: http://www.sa.infn.it/NeuralGroup/ Dr. Silvia Scarpetta Dipartimento di Fisica "E.R. Caianiello" Universit degli Studi di Salerno Via S. Allende 84081 Baronissi (SA) Italy Tel. +39 089 965318 Web page: http://www.sa.infn.it/silvia.scarpetta From marc.vanhulle at med.kuleuven.ac.be Tue Aug 2 07:34:02 2005 From: marc.vanhulle at med.kuleuven.ac.be (Marc Van Hulle) Date: Tue, 02 Aug 2005 13:34:02 +0200 Subject: Connectionists: differential log-likelihood Message-ID: <42EF5A2A.ABD44E7B@med.kuleuven.ac.be> Dear Connectionists, Two papers on a new and unbiased information metric, called differential log-likelihood, are available. Regular information metrics such as Akaike's, BIC and GIC start from the log-likelihood but do not produce an unbiased metric. This implies that the optimal case corresponds to a minimum. The differential log-likelihood is optimal when it reaches the zero value, and a zero-crossing is much easier to locate than a minimum. Van Hulle, M.M. (2005). Differential log-likelihood for evaluating and learning Gaussian mixtures. Neural Computation, in press. Abstract We introduce a new unbiased metric for assessing the quality of density estimation based on Gaussian mixtures, called differential log-likelihood. As an application, we determine the optimal smoothness and the optimal number of kernels in Gaussian mixtures. Furthermore, we suggest a learning strategy for Gaussian mixture density estimation, and compare its performance with log-likelihood maximization for a wide range of real-world data sets. http://134.58.34.60/~marc/NC11.pdf -- Van Hulle, M.M. (2005). Mixture density modeling, Kullback-Leibler divergence, and differential log-likelihood. Signal Processing (Special Issue on Information Theoretic Signal Processing), J. Principe & D. Erdogmus (Eds.), 85(5), 951-963. http://134.58.34.60/~marc/SP.pdf -- Marc M. Van Hulle K.U.Leuven Laboratorium voor Neurofysiologie Faculteit Geneeskunde (Medical School) Campus Gasthuisberg Bus 801 Herestraat 49 B-3000 Leuven Belgium Phone: + 32 16 345961 Fax: + 32 16 345960 E-mail: marc.vanhulle at med.kuleuven.ac.be URL: http://simone.neuro.kuleuven.ac.be From terry at salk.edu Wed Aug 3 16:58:24 2005 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 03 Aug 2005 13:58:24 -0700 Subject: Connectionists: NEURAL COMPUTATION 17:9 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 17, Number 9 - September 1, 2005 REVIEW The Cocktail Party Problem Simon Haykin and Zhe Chen NOTES Edgeworth Approximation of Multivariate Differential Entropy Marc M. Van Hulle Perfect Fault Tolerance of the /n-k-n/ Network Elko B. Tchernev, Rory G. Mulvaney and Dhananjay S. Phatak On the Slow Convergence of EM and VBEM in Low-Noise Linear Models Kaare Brandt Petersen, Ole Winther and Lars Kai Hansen LETTERS Analyzing Functional Connectivity Using a Network Likelihood Model of Ensemble Neural Spiking Activity Murat Okatan, Matthew A. Wilson, and Emery N. Brown Data-Robust Tight Lower Bounds to the Information Carried by Spike Times of a Neuronal Population G. Pola, R. S. Petersen, A. Thiele, M. P. Young and S. Panzeri Fluctuation-Disspation Theorem and Models of Learning Ilya Nemenman Correlated Firing in a Feedforward Network with Mexican-Hat-Type Connectivity Kosuke Hamaguchi, Masato Okada, Michiko Yamana, and Kazuyuki Aihara Supervised Learning in a Recurrent Network of Rate-Model Neurons Exhibiting Frequency Adaptation Pierre A. Fortier, Emmanuel Guigon and Yves Burnod Learning Bounds for Kernel Regression Using Effective Data Dimensionality Tong Zhang ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2005 - VOLUME 17 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $114 $54 $57.78 Individual $100 $107.00 $143 $90 $96.30 Institution $680 $727.60 $734 $612 $654.84 * 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 wduch at phys.uni.torun.pl Fri Aug 5 20:42:39 2005 From: wduch at phys.uni.torun.pl (wduch@phys.uni.torun.pl) Date: Sat, 06 Aug 2005 02:42:39 +0200 Subject: Connectionists: Free tutorials at the ICANN conference, Sept. 11, Warsaw, Poland Message-ID: <1123288959.42f4077f1dc4c@poczta.umk.pl> We are pleased to announce the following tutorials, free for all registered participants of the ICANN-2005, http://www.ibspan.waw.pl/ICANN-2005/index.html 1. An Introduction to Bayesian Methods, by Christopher Bishop (Microsoft, Cambridge, UK) 2. Spiking neural networks, by Hans-Heinrich Bothe (Technical University of Denmark) 3. Evolving Connectionist Systems: Biological Principles, Models and Applications by Nik Kasabov (Knowledge Engineering and Discovery Research Institute, Auckland University of Technology, New Zealand) 4. Networks of Evolutionary Processors: A Survey by Carlos Martin-Vide (Rovira i Virgili University, Spain) 5. Recurrent neural networks: state of the art, by Juergen Schmidhuber (IDSIA, Switzerland and TU Munich, Germany) 6. Through Attention to Creating Conscious Machines, by John Gerald Taylor (King's College London, UK) They will be presented on September 11, 2005 (Sunday) at the Gromada Airport Hotel, Warsaw, Poland (the conference venue). From auke.ijspeert at epfl.ch Tue Aug 9 10:52:34 2005 From: auke.ijspeert at epfl.ch (Auke Ijspeert) Date: Tue, 09 Aug 2005 16:52:34 +0200 Subject: Connectionists: AMAM2005: call for participation and last-minute-results abstracts Message-ID: <42F8C332.2070708@epfl.ch> AMAM2005, Sept 25-30 2005, Ilmenau, http://www.tu-ilmenau.de/amam Dear Connectionist colleagues, Researchers working on any aspect related to the adaptive control of movement and locomotion in animals and robots might be interested in participating to AMAM2005, the Third International Symposium on Adaptive Motion in Animals and Machines. The symposium will take place at the Technische Universit?t Ilmenau, Germany, from September 25th to September 30th, 2005. The two previous symposia in Montreal and Kyoto, which brought together researchers in neurobiology, biomechanics, neural computation, and robotics, were very exciting and fruitful events. The program is now fixed, see http://www.tu-ilmenau.de/amam , but there is the opportunity to submit an abstract to a special *"last-minute-results" poster session*, see the CFP below, and* to bring robots to participate to public robot demonstrations*, see http://www.tu-ilmenau.de/amam (Robot Data Sheet link). To hold keynotes ("Basics & Advances in ...") we have confirmations from the following colleagues: Biology & Mechanics: *Roy Ritzmann*, Movement Through Complex Terrain: From Animals to Robots and Back I *Martin S. Fischer*, Movement Through Complex Terrain: From Animals to Robots and Back II Biology & Control: *Sten Grillner*, The Neural control of vertebrate locomotion - from ion channels to behavior *Auke Jan Ijspeert*, Pattern generators in the central nervous system: numerical models and applications to robotics Bionics/Biomimetics: *Werner Nachtigall*, Technical Biology - Basis for Bionical Realization (supported by BioKoN) *Rudolf Bannasch*, The German Bionics Competence Network "BioKoN" (supported by BioKoN) *Rolf Pfeifer*, Morphological computation: connecting body, brain and environment (supported by BioKoN) Robotics & Control: *Ruediger Dillmann*, Title to be announced *Vladimir V. Beletsky*, Regular and Chaotic Dynamics of Two-Legged Walking Robotics & Mechanics: *Martin Buehler*, Legged Robots Step Outside *Fumiya Iida*, Cheap Design Approach to Adaptive Behavior: Walking and Sensing through Body Dynamics With best regards, Hartmut Witte, Hiroshi Kimura, and Auke Ijspeert on behalf of the AMAM organizing committee. _ Call for last-minute-results posters:_ Researchers are invited to submit a one-page abstract describing their most recent results/developments. The call for abstracts is open to all (i.e. to people both with and without an accepted paper at the conference). There is a limit of one abstract/poster per participant as a first author. After a quick review by the conference chairs, authors of accepted abstracts will be invited to present their results as a poster in a special Last-Minute-Results session (short oral spotlight presentation + poster presentation). Note that the abstracts will not be included in the conference proceedings, but will be included in a special booklet. The* deadline for submission is September 10^th *, but can be submitted at any time from now on. We will try to review submissions and return a notification within one week. Please email your abstracts to amam at tu-ilmenau.de From oby at cs.tu-berlin.de Wed Aug 10 08:19:38 2005 From: oby at cs.tu-berlin.de (Klaus Obermayer) Date: Wed, 10 Aug 2005 14:19:38 +0200 (MEST) Subject: Connectionists: PhD and Postdoc positions Message-ID: Can you please post this job announcement on the list? Thanks and all the best Klaus =========================================================================== The Department for Electrical Engineering and Computer Science of the Technial University of Berlin (Germany) solicits applications for 1 Postdoctoral Position (BAT Ib) 1 Graduate Student Position (BAT IIa) (starting date October 1st 2005). The research tasks are in the field of machine learning / data mining and include the application and development of machine learning methods for developing adaptive and personalized e-commerce websites. Both positions are part of the joint project 'Cybersmart' in cooperation with several industrial and academic partners. Requirements: Candidates should hold a recent PhD-degree (postdoctoral position) or a Diplom or Master degree (graduate student position) in computer science, physics, mathematics, statistics, or related fields. Strong mathematical and programming skills, and research experience in the field of machine learning are required. The postdoctoral position will is also involved in the coordination of the project. The positions are initially for 2 years. Salary is according to the German BAT Ib (postdoctoral position) resp. BAT IIa (graduate student position) level. Application material (CV, list of publications, copies of certificates, summary of the doctoral resp. diploma thesis and two letters of reference) should be sent within two weeks after the release of this announcement to Prof. Dr. Klaus Obermayer, FR 2-1, Technische Universit?t Berlin, Franklinstra?e 28/29, 10587 Berlin, email: oby at cs.tu-berlin.de. For more information see: http://ni.cs.tu-berlin.de/. TUB seeks to increase the proportion of women and particularly encourages women to apply. Women will be preferred given equal qualifications. Disabled persons will be preferred given equal qualifications. From marc at memory.syr.edu Fri Aug 12 11:19:24 2005 From: marc at memory.syr.edu (Marc Howard) Date: Fri, 12 Aug 2005 11:19:24 -0400 (EDT) Subject: Connectionists: Tenure-track faculty position at Syracuse Univ Message-ID: Greetings Connectionists, Thought this opening might be of interest to some readers of this list. Thanks, marc -- marc at memory.syr.edu Marc Howard, Ph.D. Syracuse University Department of Psychology http://memory.syr.edu/ (315) 443-1864 ------------------------------------------------------------------------------ The Psychology Department of Syracuse University invites applications for a tenure-track position in Cognitive Psychology. Candidates with post-doctoral experience and with expertise in quantitative modeling or in the use of psychophysiological measures in any area of cognitive psychology (e.g., memory, learning, attention, emotion, perception, cognitive neuroscience) are encouraged to apply. Responsibilities of the position include 1) establishing or continuing to build a program of research in one of the areas mentioned; and 2) research supervision and teaching at the undergraduate and graduate level (e.g., courses in brain and cognition, quantitative methods, and/or in areas of research specialization). This position is at the assistant professor level although candidates at the advanced assistant level are encouraged to apply. Candidates with interest and expertise in teaching a graduate level course in statistics, and with substantive and statistical expertise that complement areas of strength in the department (social problems, cognitive aging, health, and child/school), are highly desirable. Collaborative research is a signature of the deparment and institution. Candidates who are interested in cross-area as well as interdisciplinary teaching and research are encouraged to apply. Successful candidates will evidence a commitment to involving students in an active research program, high quality scholarship and the potential for extramural funding. Additional information about the department may be found at http://psychweb.syr.edu. Submit a letter of application (including statements of teaching and research interests), a curriculum vita, three letters of recommendation, and evidence of teaching effectiveness to the Chair of the Cognitive Search Committee; 430 Huntington Hall, Department of Psychology, Syracuse University, Syracuse, NY 13244-2340. Applications received by November 1 will receive full consideration, with subsequent applications considered until the position is filled. Syracuse University is an equal opportunity, affirmative-action employer. The Psychology Department is committed to enhancing the diversity of its faculty and especially encourages applications from women, members of minority groups, and individuals with disabilities. From nicolas.lewy at unine.ch Mon Aug 15 11:03:04 2005 From: nicolas.lewy at unine.ch (LEWY Nicolas) Date: Mon, 15 Aug 2005 17:03:04 +0200 Subject: Connectionists: Paper on model of spoken word recognition in French Message-ID: <4300AEA8.6010804@unine.ch> The following article (written in French) on an interactive activation model of spoken word recognition in French has just been published on-line. Lwy, N., Grosjean, F., Grosjean, L., Racine, I. & Yersin, C. (2005). "Un modle psycholinguistique informatique de la reconnaissance des mots dans la chane parle du franais". Journal of French Language Studies, Volume 15, Issue 1, pages 2548. For (currently free) access via your WWW browser, use this URL: http://journals.cambridge.org/action/displayIssue?jid=JFL&volumeId=15&issueId=01 Abstract: Nous proposons un nouveau modle psycholinguistique informatique du franais. Le modle, intitul FN5, porte sur la reconnaissance de mots parls, prsents en isol (dterminant, adjectif antpos, substantif) ou en suites de deux mots (dterminant et substantif, adjectif antpos et substantif). Grce un lexique de plus de 17 000 mots, une architecture connexionniste localiste, et des mcanismes dvelopps spcifiquement (processeur de position, groupements de connexions, et point d'isolation), le modle peut simuler, de manire squentielle ou simultane, certains effets propres au mot isol (frquence, longueur, homophonie) ou une suite de mots (coupure lexicale, enchanement, liaison, effacement du schwa). De plus, le modle tient compte de certaines diffrences qui existent entre le franais standard et le franais de Suisse romande (nombre de voyelles, dure vocalique en fin de mot, et statut du schwa). Nous dcrivons ce modle, prsentons son interface graphique, et l'illustrons avec des exemples de simulation. Thank you for your interest / Merci de votre intrt, Nicolas Lwy Laboratoire de traitement du langage et de la parole (Language and Speech Processing Laboratory) Universit de Neuchtel, Avenue du Premier-Mars 26 CH-2000 Neuchtel, Switzerland email: nicolas.lewy at unine.ch From jcortes at onsager.ugr.es Tue Aug 16 03:04:16 2005 From: jcortes at onsager.ugr.es (J.M. Cortes) Date: Tue, 16 Aug 2005 09:04:16 +0200 (CEST) Subject: Connectionists: paper available on fast presynaptic noise on ANN Message-ID: Dear all, I would like to announce the following paper, that can be downloaded from arxiv.org/pdf/q-bio.NC/0508013 and that will be published in Neural Computation this year 2005. Effects of fast presynaptic noise in attractor neural networks J. M. Cortes, J. J. Torres, J. Marro, P. L. Garrido, H. J. Kappen We study both analytically and numerically the effect of presynaptic noise on the transmission of information in attractor neural networks. The noise occurs on a very short-time scale compared to that for the neuron dynamics and it produces short-time synaptic depression. This is inspired in recent neurobiological findings that show that synaptic strength may either increase or decrease on a short-time scale depending on presynaptic activity. We thus describe a mechanism by which fast presynaptic noise enhances the neural network sensitivity to an external stimulus. The reason for this is that, in general, the presynaptic noise induces nonequilibrium behavior and, consequently, the space of fixed points is qualitatively modified in such a way that the system can easily scape from the attractor. As a result, the model shows, in addition to pattern recognition, class identification and categorization, which may be relevant to the understanding of some of the brain complex tasks. Jesus M. Cortes, PostDoctoral Researcher SNN, Radboud University Nijmegen The Netherlands Tel. +31 243614230 Fax. +31 243541435 From nicolas.lewy at unine.ch Mon Aug 15 14:25:22 2005 From: nicolas.lewy at unine.ch (LEWY Nicolas) Date: Mon, 15 Aug 2005 20:25:22 +0200 Subject: Connectionists: Paper on model of spoken word recognition in French (* corrected *) Message-ID: <4300DE12.5040701@unine.ch> (* The previous version of this message has unfortunately been stripped of accented characters. Please read this corrected message instead *) The following article (written in French) on an interactive activation model of spoken word recognition in French has just been published on-line. Lewy, N., Grosjean, F., Grosjean, L., Racine, I. & Yersin, C. (2005). "Un modele psycholinguistique informatique de la reconnaissance des mots dans la chaine parlee du francais". Journal of French Language Studies, Volume 15, Issue 1, pages 2548. For (currently free) access via your WWW browser, use this URL: http://journals.cambridge.org/action/displayIssue?jid=JFL&volumeId=15&issueId=01 Abstract: Nous proposons un nouveau modele psycholinguistique informatique du francais. Le modele, intitule' FN5, porte sur la reconnaissance de mots parles, presentes en isole (determinant, adjectif antepose, substantif) ou en suites de deux mots (determinant et substantif, adjectif antepose et substantif). Grace a un lexique de plus de 17 000 mots, a une architecture connexionniste localiste, et a des mecanismes developpes specifiquement (processeur de position, groupements de connexions, et point d'isolation), le modele peut simuler, de maniere sequentielle ou simultanee, certains effets propres au mot isole (frequence, longueur, homophonie) ou a une suite de mots (coupure lexicale, enchainement, liaison, effacement du schwa). De plus, le modele tient compte de certaines differences qui existent entre le francais standard et le francais de Suisse romande (nombre de voyelles, duree vocalique en fin de mot, et statut du schwa). Nous decrivons ce modele, presentons son interface graphique, et l'illustrons avec des exemples de simulation. Thank you for your interest / Merci de votre interet, Nicolas Lewy Laboratoire de traitement du langage et de la parole (Language and Speech Processing Laboratory) Universite de Neuchatel, Avenue du Premier-Mars 26 CH-2000 Neuchatel, Switzerland email: nicolas.lewy at unine.ch From oza at email.arc.nasa.gov Mon Aug 15 19:05:26 2005 From: oza at email.arc.nasa.gov (Nikunj Oza) Date: Mon, 15 Aug 2005 16:05:26 -0700 Subject: Connectionists: CFP: Information Fusion Journal - Special Issue on Applications of Ensemble Methods Message-ID: <43011FB6.6050809@email.arc.nasa.gov> APOLOGIES FOR MULTIPLE COPIES Call for papers for a special issue of Information Fusion An International Journal on Multi-Sensor, Multi-Source Information Fusion An Elsevier Publication On APPLICATIONS OF ENSEMBLE METHODS Editor-in-Chief: Dr. Belur V. Dasarathy, FIEEE d.belur at elsevier.com http://belur.no-ip.com Guest Editors: Nikunj C. Oza, Kagan Tumer The Information Fusion Journal is planning a special issue devoted to Applications of Ensemble Methods in Machine Learning and Pattern Recognition. Ensembles, also known as Multiple Classifier Systems (MCSs) and Committee Classifiers, were originally motivated by the desire to avoid relying on just one learned model when only a small amount of training data is available. Because of this, most studies on ensembles have evaluated their new algorithms on relatively small datasets; most notably, datasets from the University of California, Irvine (UCI) Machine Learning Repository. However, modern data mining problems raise a variety of issues very different from the ones ensembles have traditionally addressed. These new problems include too much data; data that are distributed, are noisy, and represent changing environments; and performance measures different from the standard accuracy measurements; among others. The aim of this issue is to examine the different applications that raise these modern data mining problems, and how current and novel ensemble methods aid in solving these problems. Manuscripts (which should be original and not previously published or presented even in a more or less similar form under any other forum) covering new applications as well as the theories and algorithms of ensemble learning algorithms developed to address these applications are invited. Contributions should be described in sufficient detail to be reproducible on the basis of the material presented in the paper. Topics appropriate for this special issue include, but are not limited to: ? Innovative applications of ensemble methods. ? Novel algorithms that address unique requirements (for example, different performance measures or running time constraints) of an application or a class of applications. ? Novel theories developed under assumptions unique to an application or a class of applications. ? Novel approaches to distributed model fusion. Manuscripts should be submitted electronically online at http://ees.elsevier.com/inffus (The corresponding author will have to create a user profile if one has not been established before at Elsevier.) Please also send without fail an electronic copy to oza at email.arc.nasa.gov (PDF format preferred), Guest Editors Nikunj C. Oza and Kagan Tumer NASA Ames Research Center Mail Stop 269-3 Moffett Field, CA 94035-1000, USA Deadline for Submission: November 30, 2005 -- -------------------------------------- Nikunj C. Oza, Ph.D. Tel: (650)604-2978 Research Scientist Fax: (650)604-4036 NASA Ames Research Center E-mail: oza at email.arc.nasa.gov Mail Stop 269-3 Web: http://ic.arc.nasa.gov/people/oza Moffett Field, CA 94035-1000 USA From wduch at phys.uni.torun.pl Tue Aug 16 08:28:41 2005 From: wduch at phys.uni.torun.pl (wduch@phys.uni.torun.pl) Date: Tue, 16 Aug 2005 14:28:41 +0200 Subject: Connectionists: Workshop: Creating an Artificial Brain, Sept 15, Torun, Poland Message-ID: <1124195321.4301dbf9339b2@poczta.umk.pl> ‘Creating an Artificial Brain’ Organisers: W Duch & JG Taylor Date: Sep 15, 2005, after ICANN 2005 Place: Nicolaus Copernicus University, Torun, Poland There is an upsurge of interest in the attempt to build an artificial brain. This is for several reasons: i) The enormous increase of computing power over the last decade or so; ii) The increased understanding of the brain arising from developments in brain imaging and single cell recordings and the associated use of probing psychological paradigms; iii) The increased demands from industry in terms of the creation of autonomous agents with some form of cognitive powers, even up to that of ‘conscious’ machines. An important aspect of this general advance is that guidance is being taken from the higher processing power of the brain, to help produce systems with a similar or even greater power. In so doing an overlap is now growing between machine intelligence per se and computational neuroscience, with projects being guided by global brain processing as observed by neuroscience. Already several such projects, using clusters of up to ten thousand nodes are being pursued; with the largest such cluster of 150,000 nodes indicating what is now available. This workshop proposes to explore the issues raised in following such an avenue: 1) The nature of the global brain itself, and the data relevant to help guide us forward; 2) The problems of the level at which to base any such approach, in terms of neuron complexity, connectivity levels, number of separate modules, learning rules, use of neuro-modulators, range of brain science data to be explained or incorporated in any model; 3) The problems of size and speed of the resulting computation, and whether to aim for real-time or off-line learning and responses 4) How to incorporate the faculties of attention, emotion and memory in an input-output system able to develop concepts and to ultimately be able to ‘think for itself’. Applications are welcome for proposed talks in the workshop (which will last one day). Please send suggestions (title, author co-ordinates and abstract) to the organizers at john.g.taylor at kcl.ac.uk and ASWDuch at ntu.edu.sg. We look forward to seeing you at the Workshop! For details on the venue and organization please see: http://www.ibspan.waw.pl/ICANN-2005/workshops.html From cindy at bu.edu Wed Aug 17 12:18:04 2005 From: cindy at bu.edu (Cynthia Bradford) Date: Wed, 17 Aug 2005 12:18:04 -0400 Subject: Connectionists: Neural Networks 18(7) 2005 Message-ID: <200508171618.j7HGI4UN002931@kenmore.bu.edu> NEURAL NETWORKS 18(7) Contents - Volume 18, Number 7 - 2005 ------------------------------------------------------------------ *****Psychology and Cognitive Science***** Elman topology with sigma-pi units: An application to the modeling of verbal hallucinations in schizophrenia Juan C. Valle-Lisboa, Florencia Reali, Hector Anastasia, and Eduardo Mizraji Linear recursive distributed representations Thomas Voegtlin and Peter F. Dominey *****Mathematical and Computational Analysis***** Generalized hamming networks and applications Konstantinos Koutroumbas and Nicholas Kalouptsidis Simultaneous L-p approximation order for neural networks Zong-Ben Xu and Fei-Long Cao Stochastic complexities of reduced rank regression in Bayesian estimation Miki Aoyagi and Sumio Watanabe *****Engineering and Design***** Biological engineering applications of feedforward neural networks designed and parameterized by genetic algorithms Konstantinos P. Ferentinos Sensitivity analysis applied to the construction of radial basis function networks D. Shi, D.S. Yeung, and J. Gao A robust classifier combined with an auto-associative network for completing partly occluded images Takashi Takahashi and Takio Kurita Data-partitioning using the Hilbert space filling curves: Effect on the speed of convergence of Fuzzy ARTMAP for large database problems Jose Castro, Michael Georgiopoulos, Ronald Demara, and Avelino Gonzalez *****Technology and Applications***** Wavelet neural network classification of EEG signals by using AR model with MLE preprocessing Abdulhamit Subasi, Ahmet Alkan, Etem Koklukaya, and M. Lemal Kiymik Using ICA for removal of ocular artifacts in EEG recorded from blind subject Arthur Flexer, Herbert Bauer, Jurgen Pripfl, and Georg Dorffner *****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 Y 11,000 (plus Y 2,000 enrollment fee) ---------------------------------------------------------------------------- membership without $30 SEK 200 not available Neural Networks to non-student (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 2810 Crossroads Drive, Suite 3800 Madison WI 53718 USA 608 443 2461, ext. 138 (phone) 608 443 2474 (fax) srees at reesgroupinc.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 JNNS Secretariat c/o Fuzzy Logic Systems Institute 680-41 Kawazu, Iizuka Fukuoka 820-0067 Japan 81 948 24 2771 (phone) 81 948 24 3002 (fax) jnns at flsi.cird.or.jp http://www.jnns.org/ ---------------------------------------------------------------------------- From M at meeter.nl Thu Aug 18 07:51:32 2005 From: M at meeter.nl (M@meeter.nl) Date: Thu, 18 Aug 2005 13:51:32 +0200 (CEST) Subject: Connectionists: papers on models of the hippocampus Message-ID: <3114.130.37.96.138.1124365892.squirrel@webmail.internl.net> Connectionists may be interested in the following four papers on the hippocampus and memory. Of all papers final drafts are available at www.cs.vu.nl/~cogsci/cogpsy/meeter. * Meeter, M., Talamini, L.M. & Murre, J.M.J. (2004). Mode shifting between storage and recall based on novelty detection in oscillating hippocampal circuits. Hippocampus, 14, 722-741. Presents a relatively low-level model of the hippocampus. Building on ideas of Hasselmo, effects of novelty and Acetylcholine (ACh) are explored. In particular, the model shows that the relatively slow time course of ACh's effects in the hippocampus are consistent with the time scale of learning. http://www3.interscience.wiley.com/cgi-bin/fulltext/107633281/PDFSTART * Meeter, M. Talamini, L.M., Schmitt, J.A.J., & Riedel, W.J. (in press). Effects of 5-HT on memory and the hippocampus: model and data. Neuropsychopharmacology. The model of Meeter et al. (2004) applied to serotonin's effects on memory. http://www.cs.vu.nl/~cogsci/cogpsy/meeter/p_publications_files/meeteretal-npp-web-jul05.pdf * Talamini, L.M., Meeter, M., Murre, J.M.J., Elvevag, B. & Goldberg, T.E. (2005). Reduced parahippocampal connectivity produces schizophrenia-like deficits in simulated neural circuits with reduced parahippocampal connectivity. Archives of General Psychiatry, 62, 485-493. A more abstract model of the medial temporal lobe and episodic memory is presented. The model suggests that abnormalities in entorhinal connectivity may cause the episodic memory deficits seen in schizophrenia (the rather silly title was a last-minute "improvement" by the journal). http://archpsyc.ama-assn.org/cgi/reprint/62/5/485 * Meeter, M., Myers, C.E., & Gluck, M.A. (2005). Integrating incremental learning and episodic memory models of the hippocampal region. Psychological Review, 112, 560-585. Theories of episodic memory and of incremental learning make mutually exclusive claims on the hippocampus. We present a model of the medial temporal lobe that reconciles these claims, and incorporates a new theory of familiarity. http://www.cs.vu.nl/~cogsci/cogpsy/meeter/p_publications_files/meeteretal-psychrev-web-dec04.pdf Best, Martijn Meeter ------------------------------ Dr. M. Meeter Dept. of Cognitive Psychology Vrije Universiteit Amsterdam The Netherlands ------------------------------ From pli at richmond.edu Thu Aug 18 15:30:23 2005 From: pli at richmond.edu (Ping Li) Date: Thu, 18 Aug 2005 15:30:23 -0400 (EDT) Subject: Connectionists: papers available Message-ID: <49809.141.166.178.153.1124393423.squirrel@141.166.178.153> Dear Colleague, You might be interested in the following three papers: Li, P., Farkas, I., & MacWhinney (2004). Early lexical acquisition in a self-organizing neural network. Neural Networks, 17, 1345-1362. Hernandez, A., Li, P., & MacWhinney, B. (2005). The emergence of competing modules in bilingualism. Trends in Cognitive Sciences, 9, 220-225. Li, P., & Cunningham, K. (2005). The APA Style Converter: A Web-based interface for converting articles to APA Style for publication. Behavior Research Methods, 37, 219-223. The papers may be requested for download at http://cogsci.richmond.edu/publications.html Sincerely Ping Li From rodrigo at neuron.ffclrp.usp.br Thu Aug 18 16:23:00 2005 From: rodrigo at neuron.ffclrp.usp.br (Rodrigo Freire Oliveira) Date: Thu, 18 Aug 2005 17:23:00 -0300 Subject: Connectionists: Second Call: LASCON*2006 Message-ID: <4304EE24.6020006@neuron.ffclrp.usp.br> Second Call - LASCON*2006 the I Latin American School on Computational Neuroscience LASCON 2006 http://neuron.ffclrp.usp.br/lascon/home.htm January 15-28 2006 University of Sao Paulo Ribeirao Preto, SP Brazil Faculty David Beeman, University of Colorado, Boulder, CO, USA James Bower, University of Texas, San Antonio, TX, USA Kim Blackwell, George Mason University, Fairfax, VA, USA Michael Hines, Yale University, New Haven, CT, USA Michael Hasselmo, Boston University, Boston, MA, USA Dieter Jaeger, Emory University, Atlanta, GA, USA Roland Koberle, University of Sao Paulo, Sao Carlos, SP, Brazil Marcelo Mazza, University of Sao Paulo, Ribeirao Preto, SP, Brazil Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Reynaldo Pinto, University of Sao Paulo, Sao Paulo, SP, Brazil Arnd Roth, University College, London, UK Michael Vanier, CALTECH, Pasadena, CA, USA Charles Wilson, University of Texas, San Antonio, TX, USA Organizer Antonio Roque, University of Sao Paulo, Ribeirao Preto, SP, Brazil Co-organizer Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Scientific Committee David Beeman, University of Colorado, Boulder, CO, USA James Bower, University of Texas, San Antonio, TX, USA Rodrigo Oliveira, University of Sao Paulo, Ribeirao Preto, SP, Brazil Deadline for student application: September 9, 2005 (Friday) The I Latin American School on Computational Neuroscience (LASCON 2006) aims at introducing advanced undergraduate and graduate students to the use of methods for detailed modeling of neurons and neural circuits, based on the Hodgkin-Huxley formalism, the cable equation and the compartmental modeling technique. The use of these methods will be illustrated with the development and investigation of numerical simulations with the programs GENESIS and NEURON. The school is divided in two weeks, the first one for theoretical lectures and hands on tutorials, and the second one for invited lectures and the development of project works by the students. Students will follow a highly demanding schedule of morning lectures followed by afternoon and evening computational laboratory sessions. The School will be held at the campus of the University of S?o Paulo at Ribeir?o Preto (a city about 313 km north of Sao Paulo). Course applicants should be fluent in English (written/spoken) and have a solid background in life and/or hard sciences (some experience in computer programming is also desirable). Applications are welcome and should be made by using the application form on the school web page (http://neuron.ffclrp.usp.br/lascon/home.htm). The number of students is limited to 20 and priority will be given to students from Latin American countries. Under exceptional circumstances, students from other areas of the world could be accepted as well. Costs for accommodation and meals will be covered by the school organization. In exceptional cases, limited funding is available to partly cover travel expenses. More information on the school can be found at http://neuron.ffclrp.usp.br/lascon/home.htm From mo at ecs.soton.ac.uk Mon Aug 22 12:30:03 2005 From: mo at ecs.soton.ac.uk (Manfred Opper) Date: Mon, 22 Aug 2005 17:30:03 +0100 Subject: Connectionists: postdoctoral position in Southampton (ERRATUM) Message-ID: <4309FD8B.7306300B@ecs.soton.ac.uk> In our job posting for the project "Inference in Complex Stochastic Dynamic Environmental Models" at the ISIS group in Southampton we gave the WRONG link http://www.jobs.soton.ac.uk/adminweb/jsp/jobs/sJobview.jsp?function=View&id=04R0842 for the application details. The CORRECT link is http://www.jobs.soton.ac.uk/adminweb/jsp/jobs/sJobview.jsp?function=View&id=04R0854 Could any person who has applied with the WRONG REF NUMBER 04R0842 please send me an email (mo at ecs.soton.ac.uk), so that we can redirect their applications? Sorry for the mistake Manfred Opper From gbrown at cs.man.ac.uk Wed Aug 24 05:15:08 2005 From: gbrown at cs.man.ac.uk (Gavin Brown) Date: Wed, 24 Aug 2005 10:15:08 +0100 Subject: Connectionists: PhD studentship at Manchester Message-ID: <430C3A9C.7030500@cs.man.ac.uk> ======================================================================= A three-year, fully-funded PhD studentship is available in the AI group at the University of Manchester, UK. The following project is suggested, though additional studentships are available in the general area of Machine Learning. Application details below. ============= "Interactions between Learning Classifier Systems and Ensemble Systems" For best generalisation ability, an ensemble of predictors must exhibit a "diversity" of errors---for majority voting ensembles, how to define this concept remains an open research question. The problem can be seen as one of credit assignment, judging how much a particular predictor contributes to the overall classification error. In spite of the shortage of theoretical results, ensembles typically show significant performance improvements over single predictors. Learning Classifier Systems (LCS) use evolutionary computation techniques to generate groups of rules---the decisions of these rules are combined by a weighted voting mechanism. Recent works in the literature have extensively studied credit assignment in LCS, providing significant insights---in spite of this and other contributions, LCS remain underused in the wider ML community. The aim of the project is to transfer theoretical results and empirical techniques between these two fields, providing both with new insights into how to understand their learning systems, and new mechanisms for solving their datasets. A priority is to replace current heuristics with theoretically justified mechanisms, so there will be a significant (but not exclusive) emphasis on the mathematical elements of the project. ============= Applicants should hold a first or upper second class degree in computer science, mathematics, physics, or similarly numerate disciplines. Additional research experience such as during a Master's degree will be seen as a significant advantage. A solid background in programming (e.g. Java/C++) is also essential. For further details and informal discussions, email Dr Gavin Brown: Email: gavin.brown at manchester.ac.uk Web : http://www.cs.man.ac.uk/~gbrown/ Those interested in the position should send detailed CVs, including: 1) full academic history, 2) description of motivations and research interests, 3) contact details for two referees, either by email (preferable) or to the following address: Dr Gavin Brown University of Manchester School of Computer Science, Kilburn Building, Manchester, M13 9LP ======================================================================= From terry at salk.edu Thu Aug 25 14:27:26 2005 From: terry at salk.edu (Terry Sejnowski) Date: Thu, 25 Aug 2005 11:27:26 -0700 Subject: Connectionists: NEURAL COMPUTATION 17:10 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 17, Number 10 - October 1, 2005 NOTE Gradient-Based Adaptation of General Gaussian Kernels Tobias Glasmachers and Christian Igel LETTERS Learning Only When Necessary: Better Memories of Correlated Patterns in Networks with Bounded Synapses Walter Senn and Stefano Fusi Coding of Temporally Varying Signals in Networks of Spiking Neurons with Global Delayed Feedback Naoki Masuda, Brent Doiron, Andre Longtin and Kazuyuki Aihara Attention-Gated Reinforcement Learning of Internal Representations for Classification Pieter R. Roelfsema and Arjen van Ooyen Computing with Continuous Attractors: Stability and Online Aspects Si Wu and Shun-ichi Amari Optimal Signal Estimation in Neuronal Models Petr Lansky and Priscilla E. Greenwood Finite State Automata Resulting from Temporal Information Maximization and a Temporal Learning Rule Thomas Wennekers and Nihat Ay Fast Access to Concepts in Concept Lattices via Bidirectional Associative Memories Rohana Rajapakse and Michael Denham ----- ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2005 - VOLUME 17 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $64.20 $114 $54 $57.78 Individual $100 $107.00 $143 $90 $96.30 Institution $680 $727.60 $734 $612 $654.84 * 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 Wael.El-deredy at manchester.ac.uk Thu Aug 25 07:33:23 2005 From: Wael.El-deredy at manchester.ac.uk (Wael El-Deredy) Date: Thu, 25 Aug 2005 12:33:23 +0100 Subject: Connectionists: PhD studentship in graphical models Message-ID: <20050825123323281.00000001288@8VJ9J1J> Bayesian networks for knowledge discovery and collaborative filtering EPSRC CASE Studentship (see new eligibility criteria below) A PhD studentship is available for three years to develop and evaluate graphical models for collaborative filtering. Probabilistic recommender systems capable of hidden variables underpinning preference, choice and purchase behaviour will be developed within the Bayesian belief networks framework. The project will involve close interaction with industrial sponsors and will be supervised by Professor Paulo Lisboa at the school of Computing and Mathematical Sciences - Liverpool John Moores University and Dr. Wael El-Deredy at School of Psychological Sciences, University of Manchester. The successful candidates will receive an enhanced EPSRC stipend. Eligible candidates* wishing to apply should send by email a supporting statement together with a CV containing names and addresses of two referees by the closing date of Monday 26th September 2005. * Candidates must satisfy the EPSRC eligibility requirements: Recent changes to the criteria have opened full CASE studentship funding to EU nationals with residence in the UK for at least 3 years, which now include periods of stay for the purpose of higher education. For further information and applying contact Professor PJG Lisboa email: p.j.lisboa at livjm.ac.uk, tel. 0151 231 2225 --------- Wael.El-Deredy at Manchester.ac.uk School of Psychological Sciences University of Manchester Zochonis Building - Oxford Road Manchester M13 9PL UK Tel. +44(0) 161 275 2566 Fax +44(0) 161 275 2685 From hdaume at ISI.EDU Tue Aug 23 14:19:01 2005 From: hdaume at ISI.EDU (Hal Daume III) Date: Tue, 23 Aug 2005 11:19:01 -0700 (PDT) Subject: Connectionists: CFP: Bayesian Methods in NLP workshop at NIPS Message-ID: ************************************************************************ CALL FOR PAPERS Bayesian Methods for Natural Language Processing Workshop at the Neural Information Processing Systems Conference (NIPS 2005) http://www.isi.edu/~hdaume/BayesNLP/ ** Submission Deadline: 21 October 2005 ** ************************************************************************ [ Apologies for multiple postings ] OVERVIEW -------- Models of natural language processing problems are often incredibly complex, and there is never enough data to properly estimate all the required parameters. This has lead to a strong need for learning techniques with built-in capacity control; most classical solutions to this problem involve largely ad-hoc smoothing techniques. The application of Bayesian learning methods to these problems could potentially result in more effective models, for which extensive cross-validation is no longer required for hyperparameter tuning or model selection. The goals of this workshop are to bring together researchers from both the Bayesian machine learning community and the natural language processing community to enable cross-fertilization of techniques, models and applications. We wish to focus on the following issues: * Statistical Models: Current Bayesian models for text have largely focused on "bag of words" style approaches, where conditional independence is assumed between words. This leads to a convenient interpretation of a document as a sequence of draws from multinomial distributions, but does not account for any of the internal structure that exists in documents and which NLP researchers are interested in. How can we build models that move beyond the bag of words assumption? What structures are useful for modeling? How can we model these structures efficiently? Can we learn these models automatically? * Applications-oriented Models: Many statistical models for text have aimed at automatically inferring implicit relationship between varied elements of documents in a corpus. How can we use such models to aid in applications? Can we develop similar models that are aimed at solving a real-world NLP task? For what NLP applications are Bayesian techniques appropriate and how can we develop models specific to these problems? CALL FOR PARTICIPATION ---------------------- We invite submission of workshop papers that discuss ongoing or completed work dealing with Bayesian techniques applied to natural language processing problems (see below for an incomplete list of possible topics). A workshop paper should be no more than six pages in the standard NIPS format. Authorship should not be blind. Please submit a paper by emailing it in Postscript or PDF format to hdaume at isi.edu with the subject line "BNLP Submission". We anticipate accepting four to six such papers for 15 minute presentation slots (exact details will be worked out shortly). Please only submit an article if at least one of the authors will be able to attend the workshop and present the work. We are especially interested in submissions from authors in the NLP community who have not previously attended a NIPS conference. If you fall into this category, please note this in your email when you submit your paper. Relevant Topics: * Models that move beyond the bag-of-words assumption * Techniques that apply to problems other than language modeling * Structure-learning techniques for language * Bayesian extensions to well-known NLP models * Application of Bayesian techniques to NLP problems * Both supervised and unsupervised techniques are welcome We also welcome position papers of at most two pages in length that discuss, with appropriate argumentation, whether or not Bayesian techniques are applicable to NLP problems and, if so, which ones. These should be submitted in the same way as standard workshop papers. These will be used to help guide discussion during panel sessions. IMPORTANT DATES --------------- 18 Aug 05 -- Call for participation 21 Oct 05 -- Paper submission deadline 4 Nov 05 -- Notification of paper acceptance 25 Nov 05 -- Survey and position paper deadlines 9/10 Dec 05 -- Workshop in Whistler RESEARCHER SURVEY ----------------- Regardless of whether you submit a paper or not, if you are a researcher in either the Bayesian learning community or the NLP community, please complete our survey (available on the web page), which will serve to guide the panel discussions at the workshop. ORGANIZATION ------------ Hal Daume III Information Sciences Institute hdaume at isi.edu http://www.isi.edu/~hdaume/ Yee Whye Teh National University of Singapore tehyw at comp.nus.edu.sg http://www.cs.berkeley.edu/~ywteh/ From M.Denham at plymouth.ac.uk Wed Aug 31 06:50:34 2005 From: M.Denham at plymouth.ac.uk (Mike Denham) Date: Wed, 31 Aug 2005 11:50:34 +0100 Subject: Connectionists: PhD position (Research Assistant) in Computational Neuroscience Message-ID: <52A8091888A23F47A013223014B6E9FE060A8504@03-CSEXCH.uopnet.plymouth.ac.uk> University of Plymouth Faculty of Science Centre for Theoretical and Computational Neuroscience Research Assistant - Three Year Fixed Term Appointment (salary range ?14,751 - ?17,601 pa) Applications are invited for a 3-year PhD position (Research Assistant) in computational neuroscience in the Centre for Theoretical and Computational Neuroscience at the University of Plymouth, UK. The post is funded as part of an EU-funded Integrated Project: "FACETS: Fast Analog Computing with Transient States in Neural Architectures". The stated objective of FACETS is to explore and exploit the yet unknown computational principles that constitute the basis of information processing in the brain. The project involves experimental neuroscience, the construction of models and analytical descriptions for neural cells and networks and the construction of very large scale neural circuits in VLSI technology. The FACETS consortium includes fifteen of the major laboratories in Europe in these areas. The PhD student in Plymouth will work specifically on the construction of computational models of neural processing of visual information in the neocortex, in collaboration with the other participants in the project. Maintaining a close level of collaboration will require short visits to the laboratories of the project collaborators. The normal UK/EU PhD tuition fees will be funded for the duration of the three-year appointment, together with a salary in the range ?14,751 - ?17,601 per annum. Applicants for the post must have a sound educational background at first degree level in a relevant subject area, with a high standard of achievement consistent with entering into a PhD programme. They should also possess a computing/mathematics background appropriate to carrying out research in computational modelling of the brain. Informal enquiries should be made to Professor Mike Denham, Centre for Theoretical and Computational Neuroscience, University of Plymouth, Drake Circus, Plymouth, PL4 8AA, UK; tel: +44 (0)1752 232547; email: mdenham at plym.ac.uk From pli at richmond.edu Wed Aug 31 11:02:59 2005 From: pli at richmond.edu (Ping Li) Date: Wed, 31 Aug 2005 11:02:59 -0400 Subject: Connectionists: cognitive neuroscience position Message-ID: <520A2447-1A30-11DA-A4AD-0011246FEB0E@richmond.edu> The University of Richmond seeks to fill a tenure track position in Neuroscience at the assistant professor level in the Department of Psychology. Although we are particularly interested in individuals with a specialization in human cognitive or behavioral neuroscience, we will consider outstanding applicants in all areas. Faculty are expected to maintain a vigorous research program, actively engage undergraduates in substantive research, attract extramural support, and commit to highly effective teaching at all levels of the undergraduate curriculum. Candidates should have completed the Ph.D. degree by the August 2006 starting date. Send vita, statements of research plan and teaching philosophy, copies of transcripts, and three letters of reference to Scott Allison, Search Coordinator, Department of Psychology, University of Richmond, VA 23173. Review of applications will commence Oct. 1, 2005 and continue until the position is filled. The University of Richmond is a highly selective private university with approximately 3500 students located on a beautiful campus six miles west of the heart of Richmond and in close proximity to the ocean, mountains, and Washington, D.C. The University of Richmond values diversity in its faculty, staff, and student body. In keeping with this commitment, our academic community strongly encourages applications from diverse candidates and candidates who support diversity. For more information please go to http://psychology.richmond.edu. ---------------------------------------------------------------------- Ping Li, Ph.D. Associate Professor Graduate Program Coordinator Department of Psychology University of Richmond Richmond, VA 23173, USA Email: pli at richmond.edu http://www.richmond.edu/~pli/ http://cogsci.richmond.edu/ Bilingualism: Language and Cognition: http://cogsci.richmond.edu/bilingualism/bilingualism.html ----------------------------------------------------------------------- From mark at paskin.org Wed Aug 31 13:42:20 2005 From: mark at paskin.org (Mark A. Paskin) Date: Wed, 31 Aug 2005 10:42:20 -0700 Subject: Connectionists: CFP: NIPS 2005 Workshop: Intelligence Beyond the Desktop Message-ID: ################################################################ CALL FOR PARTICIPATION Intelligence Beyond the Desktop a workshop at the 2005 Neural Information Processing Systems (NIPS) Conference Submission deadline: Friday, October 14, 2005 http://ai.stanford.edu/~paskin/ibd05/ ################################################################ OVERVIEW We are now well past the era of the desktop computer. Trends towards miniaturization, wireless communication, and increased sensing and control capabilities have led to a variety of systems that distribute computation, sensing, and controls across multiple devices. Examples include wireless sensor networks, multi-robot systems, networks of smartphones, and large area networks. Machine learning problems in these non-traditional settings cannot faithfully be viewed in terms of a data set and an objective function to optimize; physical aspects of the system impose challenging new constraints. Resources for computation and actuation may be limited and distributed across many nodes, requiring significant coordination; limited communication resources can make this coordination expensive. The scale and complexity of these systems often leads to large amounts of structured data that make state estimation challenging. In addition, these systems often have other constraints, such as limited power, or under-actuation, requiring reasoning about the system itself during learning and control. Furthermore, large-scale distributed systems are often unreliable, requiring algorithms that are robust to failures and lossy communication. New learning, inference, and control algorithms that address these challenges are required. This workshop aims to bring together researchers to discuss new applications of machine learning in these systems, the challenges that arise, and emerging solutions. FORMAT This one-day workshop will consist of invited talks and talks based upon submitted abstracts, with some time set aside for discussion. Our (tentative) invited speakers are: * Dieter Fox (University of Washington) * Leonidas Guibas (Stanford University) * Sebastian Thrun, (Stanford University) will speak about the machine learning algorithms used in Stanley, Stanford's entry into the DARPA Grand Challenge. CALL FOR PARTICIPATION Researchers working at the interface between machine learning and non- traditional computer architectures are invited to submit descriptions of their research for presentation at the workshop. Of particular relevance is research on the following topics: * distributed sensing, computation, and/or control * coordination * robustness * learning/inference/control under resource constraints (power, computation, time, etc.) * introspective machine learning (reasoning about the system architecture in the context of learning/inference/control) We especially encourage submissions that address unique challenges posed by non-traditional architectures for computation, such as * wireless sensor networks * multi-robot systems * large-area networks Submissions should be extended abstracts in PDF format which are no longer than three (3) pages long in 10pt or larger font. Submissions may be e-mailed to ibd-2005 at cs.cmu.edu with the subject "IBD SUBMISSION". We plan to accept four to six submissions for 25 minute presentation slots. In your submission please indicate if you would present a poster of your work (in case there are more qualified submissions than speaking slots). Call for participation: Wednesday, August 31, 2005 Submission deadline: Friday, October 14, 2005 11:59 PM PST Acceptance notification: Tuesday, November 1, 2005 Workshop: Friday, December 9, 2005 Organizers * Carlos Guestrin (http://www.cs.cmu.edu/~guestrin/) * Mark Paskin (http://paskin.org) Please direct any inquiries regarding the workshop to ibd-2005 at cs.cmu.edu. From steve at cns.bu.edu Wed Aug 31 20:29:49 2005 From: steve at cns.bu.edu (Stephen Grossberg) Date: Wed, 31 Aug 2005 20:29:49 -0400 Subject: Connectionists: watercolor illusion and neon color spreading Message-ID: The following article is now available at http://www.cns.bu.edu/Profiles/Grossberg Baingio Pinna, and Stephen Grossberg (2005). The watercolor illusion and neon color spreading: A unified analysis of new cases and neural mechanisms. Journal of the Optical Society of America A, in press. ABSTRACT Coloration and figural properties of neon color spreading and the watercolor illusion are studied using phenomenal and psychophysical observations. Coloration properties of both effects can be reduced to a common limiting condition, a nearby color transition called the "two-dot limiting case", that clarifies their perceptual similarities and dissimilarities. The results are explained by the FACADE neural model of biological vision. The model proposes how local properties of color transitions activate spatial competition among nearby perceptual boundaries, with boundaries of lower contrast edges weakened by competition more than boundaries of higher contrast edges. This asymmetry induces spreading of more color across these boundaries than conversely. The model also predicts how depth and figure-ground effects are generated in these illusions.