From akira at bstu.by Thu Oct 1 11:55:18 2009 From: akira at bstu.by (Akira Imada) Date: Thu, 1 Oct 2009 18:55:18 +0300 Subject: Connectionists: Welcome to Belarus -- Join us next June at ICNNAI-2010 --- 8 months to go. Message-ID: <007c01ca42af$9320c9b0$0d0210ac@best> Dear Connectionist, We are organizing the 5th International Conference on Neural Network and Artificial Intelligence (ICNNAI-2010) on 1-4 June 2010 in Brest, Belarus. We have already confirmed the following world top-class influential key-note guests. - Professor Shun-ich Amari (Riken, Japan) - Professor Steven Bressler (Florida Atlantic University, US) - Professor Xin Yao (University of Birmingham, UK) (Alphabetical order) And we are negotiating with yet other prestigious researchers for stimulating key-note talks. Still we have a little time to prepare for a submission, and we'd be happy to if you mark your calendar, if not yet, and sleep on a plan to attend and present your precious theory, results or whatsoever regarding the topic. For more in detail, please visit our CFP at http://icnnai.bstu.by/icnnai-2010.html. Belarus is a fairly well-kept secret land. In this lovely wonder land, with luck, you'll find something new that could not be come across elsewhere. Looking forward to seeing you at the conference. With the warmest regards. Akira Imada on behalf of the organizing committee. ---------- 8< ---------- 8< ---------- Akira Imada Prof. Ph.D. Dept. Intelligent Information Technology Brest State Technical University Moskowskaja 267, Brest, 224017 Belarus akira-i at brest-state-tech-univ.org -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091001/4b68184b/attachment.html From t.heskes at science.ru.nl Thu Oct 1 14:06:32 2009 From: t.heskes at science.ru.nl (Tom Heskes) Date: Thu, 01 Oct 2009 20:06:32 +0200 Subject: Connectionists: Neurocomputing volume 72 (issues 13-15) Message-ID: <4AC4EFA8.1060902@science.ru.nl> Neurocomputing volume 72 (issues 13-15) ------- SPECIAL PAPERS (Financial Engineering) Financial Engineering (editorial) Mu-Yen Chen Seize the (intra)day: Features selection and rules extraction for tradings on high-frequency data Marina Resta Estimating VaR in crude oil market: A novel multi-scale non-linear ensemble approach incorporating wavelet analysis and neural network Kaijian He, Chi Xie, Shou Chen, Kin Keung Lai A combination of hidden Markov model and fuzzy model for stock market forecasting Md. Rafiul Hassan Predicting investment behavior: An augmented reinforcement learning model Tetsuya Shimokawa, Kyoko Suzuki, Tadanobu Misawa, Yoshitaka Okano Fusion ANFIS models based on multi-stock volatility causality for TAIEX forecasting Ching-Hsue Cheng, Liang-Ying Wei, You-Shyang Chen Evaluation of automated-trading strategies using an artificial market K. Izumi, F. Toriumi, H. Matsui An approach to budget allocation for an aerospace company?Fuzzy analytic hierarchy process and artificial neural network Yu-Cheng Tang Effects of Financial Holding Company Act on bank efficiency and productivity in Taiwan Chei-Chang Chiou A cross model study of corporate financial distress prediction in Taiwan: Multiple discriminant analysis, logit, probit and neural networks models Tzong-Huei Lin A hybrid stock trading system for intelligent technical analysis-based equivolume charting Thira Chavarnakul, David Enke A self-organising mixture autoregressive network for FX time series modelling and prediction He Ni, Hujun Yin ------- SPECIAL PAPERS (Computational and Ambient Inteligence, IWANN 2007) Computational and ambient intelligence (editorial) Joan Cabestany, Francisco Sandoval, Alberto Prieto Parallel multiobjective memetic RBFNNs design and feature selection for function approximation problems A. Guill?n, H. Pomares, J. Gonz?lez, I. Rojas, O. Valenzuela, B. Prieto An adaptive field rule for non-parametric MRI intensity inhomogeneity estimation algorithm Maite Garc?a-Sebasti?n, Ana Isabel Gonz?lez, Manuel Gra?a A single front genetic algorithm for parallel multi-objective optimization in dynamic environments Mario C?mara, Julio Ortega, Francisco de Toro Information-theoretic feature selection for functional data classification Vanessa G?mez-Verdejo, Michel Verleysen, J?r?me Fleury Supervised data analysis and reliability estimation with exemplary application for spectral data Frank-Michael Schleif, Thomas Villmann, Matthias Ongyerth Characterization of the convergence of stationary Fokker?Planck learning Arturo Berrones A programmable spike-timing based circuit block for reconfigurable neuromorphic computing Thomas Jacob Koickal, Luiz C. Gouveia, Alister Hamilton Evidences of cognitive effects over auditory steady-state responses by means of artificial neural networks and its use in brain?computer interfaces M.-A. Lopez, Hector Pomares, Francisco Pelayo, Jose Urquiza, Javier Perez Emerging motor behaviors: Learning joint coordination in articulated mobile robots Diego Pardo, Cecilio Angulo, Sergi del Moral, Andreu Catal? Computational intelligence tools for next generation quality of service management Rafael del-Hoyo, Bonifacio Mart?n-del-Br?o, Nicolas Medrano, Julian Fern?ndez-Navajas Sine-fitting multiharmonic algorithms implemented by artificial neural networks J.R. Salinas, F. Garcia-Lagos, G. Joya, F. Sandoval Neural projection techniques for the visual inspection of network traffic ?lvaro Herrero, Emilio Corchado, Paolo Gastaldo, Rodolfo Zunino Assessing aspects of reading by a connectionist model J. Ignacio Serrano, M. Dolores del Castillo, ?ngel Iglesias, Jes?s Oliva ------- REGULAR PAPERS Stochastic stability of Markovian jumping Hopfield neural networks with constant and distributed delays Hongyang Liu, Lin Zhao, Zexu Zhang, Yan Ou Delay-interval dependent robust stability criteria for stochastic neural networks with linear fractional uncertainties P. Balasubramaniam, S. Lakshmanan, R. Rakkiyappan Improving the training time of support vector regression algorithms through novel hyper-parameters search space reductions Emilio G. Ortiz-Garc?a, Sancho Salcedo-Sanz, ?ngel M. P?rez-Bellido, Jose A. Portilla-Figueras Residual variance estimation in machine learning Elia Liiti?inen, Michel Verleysen, Francesco Corona, Amaury Lendasse Variational Bayesian learning of nonlinear hidden state-space models for model predictive control Tapani Raiko, Matti Tornio Recognition of coloured and textured images through a multi-scale neural architecture with orientational filtering and chromatic diffusion M. Ant?n-Rodr?guez, F.J. D?az-Pernas, J.F. D?ez-Higuera, M. Mart?nez-Zarzuela, D. Gonz?lez-Ortega, D. Boto-Giralda Global asymptotic stability of delayed neural networks with discontinuous neuron activations Liping Li, Lihong Huang A normal least squares support vector machine (NLS-SVM) and its learning algorithm Xinjun Peng, Yifei Wang Multi-view face recognition based on tensor subspace analysis and view manifold modeling Xinbo Gao, Chunna Tian Almost periodic solution to Cohen?Grossberg-type BAM networks with distributed delays Hongjun Xiang, Jinhua Wang, Jinde Cao Classified self-organizing map with adaptive subcodebook for edge preserving vector quantization Chao-Huang Wang, Chung-Nan Lee, Chaur-Heh Hsieh A new learning algorithm with logarithmic performance index for complex-valued neural networks R. Savitha, S. Suresh, N. Sundararajan, P. Saratchandran A class of discrete time recurrent neural networks with multivalued neurons Wei Zhou, Jacek M. Zurada Properties of variations in the periods of ring neural oscillators with noise Yo Horikawa, Hiroyuki Kitajima Block attractor in spatially organized neural networks Mario Gonz?lez, David Dominguez, Francisco B. Rodr?guez Convergence of non-autonomous discrete-time Hopfield model with delays Lifen Yuan, Zhaohui Yuan, Yigang He Some multistability properties of bidirectional associative memory recurrent neural networks with unsaturating piecewise linear transfer functions Lei Zhang, Zhang Yi, Jiali Yu, Pheng Ann Heng A fast and accurate online self-organizing scheme for parsimonious fuzzy neural networks Ning Wang, Meng Joo Er, Xianyao Meng Delay-distribution-dependent stability of stochastic discrete-time neural networks with randomly mixed time-varying delays Yang Tang, Jian-an Fang, Min Xia, Dongmei Yu Registration for 3-D point cloud using angular-invariant feature Jun Jiang, Jun Cheng, Xinglin Chen Global synchronization of delay-coupled genetic oscillators Jianlong Qiu, Jinde Cao Adaptive particle swarm optimization for CNN associative memories design Girolamo Fornarelli, Antonio Giaquinto Control Liapunov function design of neural networks that solve convex optimization and variational inequality problems Fernando A. Pazos, Amit Bhaya Solving TSP by using Lotka?Volterra neural networks Manli Li, Zhang Yi, Min Zhu Estimation of epipolar geometry by linear mixed-effect modelling Huiyu Zhou, Patrick R. Green, Andrew M. Wallace Neural network approximation for periodically disturbed functions and applications to control design Weisheng Chen, Yu-Ping Tian Stability analysis for stochastic BAM neural networks with Markovian jumping parameters Guanjun Wang, Jinde Cao, Ming Xu Synchronization analysis of coupled connected neural networks with mixed time delays Qiankun Song Neighborhood based modified backpropagation algorithm using adaptive learning parameters for training feedforward neural networks T. Kathirvalavakumar, S. Jeyaseeli Subavathi Image retrieval using nonlinear manifold embedding Can Wang, Jun Zhao, Xiaofei He, Chun Chen, Jiajun Bu Median MSD-based method for face recognition Xiaodong Li, Shumin Fei, Tao Zhang New exponentially convergent state estimation method for delayed neural networks Magdi S. Mahmoud Combining segmental semi-Markov models with neural networks for protein secondary structure prediction Niranjan P. Bidargaddi, Madhu Chetty, Joarder Kamruzzaman ------- BRIEF PAPERS MIMLRBF: RBF neural networks for multi-instance multi-label learning Min-Ling Zhang, Zhi-Jian Wang Complex wavelet based texture classification Yu-Long Qiao, Chun-Hui Zhao, Chun-Yan Song ------- JOURNAL SITE: http://www.elsevier.com/locate/neucom SCIENCE DIRECT: http://www.sciencedirect.com/science/issue/5660-2009-999279983-1483156 From viktor.jirsa at univmed.fr Thu Oct 1 14:43:55 2009 From: viktor.jirsa at univmed.fr (Viktor Jirsa) Date: Thu, 01 Oct 2009 20:43:55 +0200 Subject: Connectionists: Postdoc position in Marseille Message-ID: <4AC4F86B.20105@univmed.fr> *Post-Doctoral position available **in *** *Computational Neuroscience * at the laboratory ? Epilepsy and Cognition ? UMR 751 INSERM * * *General environment:* * * The research is focused on the uncovering the mechanisms underlying i) the construction of an epileptic brain, ii) the genesis of seizures and iii) the cognitive deficits associated with epilepsy. We use a multidisciplinary vertical approach from the gene level to behavior, including in vitro/in vivo electrophysiology; as well as a horizontal approach from animal models to epileptic patients, from the bench to the bedside. Our team has unraveled many key mechanisms (Nature Neuroscience, 1998, 1999, 2001; Neuron 2001; Science 2004; PNAS, 2005, 2008; J Neurosc 2009, Brain, 2009) and will provide a unique multidisciplinary bilingual working environment, in which basic and clinical research are fused. *Position: Brain dynamics during the construction of an epileptic network* * * *Start date*: November 01 2009 and later. *Duration*: 3 years *Funding*: ANR grant, ANTARES *P.Is. Christophe Bernard and Viktor Jirsa* * * *Research project*: We are seeking a post-doctoral fellow with a background in non-linear dynamics and computational neuroscience. We have designed a system enabling us to record simultaneously in 17 different brain regions in rats. Local field potentials are collected in various experimental conditions (different cognitive states, including spatial and non spatial memory tasks) in normal animals, and following an insult, which will lead to epilepsy. The goal of the project is to describe and characterize the trajectories of brain activities in the different experimental conditions, to determine how activities in different regions are coordinated, and how this dynamics is modified in an epileptic context. These behaviors will be mathematically modeled and computationally implemented in a large-scale network model, whose predictions will be then experimentally tested. Experimental and theoretical research will be closely coordinated and the candidate will have the possibility to learn and perform the in vivo experiments, as well as receive training in theoretical neuroscience. *Requirements*: The successful candidate will have a Ph.D., or equivalent, in a relevant discipline (Physics or Engineering with a focus on nonlinear-dynamics, computational neuroscience,?) with an excellent past record achievement. A background in neurobiology is an advantage but is not required. *Contact*: Please send CV and two letters of recommendation to Christophe Bernard (christophe.bernard at univmed.fr ) or Viktor Jirsa (viktor.jirsa at univmed.fr). From S.R.Marsland at massey.ac.nz Thu Oct 1 16:38:20 2009 From: S.R.Marsland at massey.ac.nz (Marsland, Stephen) Date: Fri, 2 Oct 2009 09:38:20 +1300 Subject: Connectionists: New book announcement Message-ID: Dear Connectionists, please note the the publication of a new undergraduate textbook in machine learning: Machine Learning: An Algorithmic Perspective (http://www.crcpress.com/product/isbn/9781420067187) Stephen Marsland, Massey University, Palmerston North, New Zealand Publication Date: April 2009 Number of Pages: 406 This is the first textbook in the new Chapman & Hall/CRC Machine Learning and Pattern Recognition Book Series, edited by Ralf Herbrich and Thore Graepel. The book presents machine learning from an algorithmic perspective suitable for computer science and engineering students, but also provides the background needed to understand how and why these algorithms work. It is supported by code in Python, which is available from: http://seat.massey.ac.nz/personal/s.r.marsland/MLBook.html Regards, Stephen -- Dr Stephen Marsland Postgraduate Director, SEAT Associate Professor, Computer Science Massey University, Palmerston North, New Zealand Email: s.r.marsland at massey.ac.nz From rsun at rpi.edu Sat Oct 3 10:39:09 2009 From: rsun at rpi.edu (Professor Ron Sun) Date: Sat, 3 Oct 2009 10:39:09 -0400 Subject: Connectionists: Cognitive Systems Research, Vol. 10, Iss. 4, 2009 Message-ID: New issue is now available: * Cognitive Systems Research Volume 10, Issue 4, Pages 297-380 (December 2009) http://www.sciencedirect.com/science/issue/ 6595-2009-999899995-1377070 NOTE: If the URLs in this email are not active hyperlinks, copy and paste the URL into the address/location box in your browser. ================= TABLE OF CONTENTS 2) Does intelligence imply contradiction? Pages 297-315 P. Frosini http://www.sciencedirect.com/science?_ob=GatewayURL&_method=citationSearch&_urlVersion=4&_origin=SDVIALERTASCII&_version=1&_uoikey=B6W6C-4TVJ3RP-1&md5=c92a34785adbc66b1bab82dc64a9401d 3) Acquisition of hierarchical reactive skills in a unified cognitive architecture Pages 316-332 Pat Langley, Dongkyu Choi, Seth Rogers http://www.sciencedirect.com/science?_ob=GatewayURL&_method=citationSearch&_urlVersion=4&_origin=SDVIALERTASCII&_version=1&_uoikey=B6W6C-4V3SY9N-1&md5=3b4bf58586cc4ce4460d2365e514ad98 4) Small worlds and Red Queens in the Global Workspace: An information- theoretic approach Pages 333-365 James F. Glazebrook, Rodrick Wallace http://www.sciencedirect.com/science?_ob=GatewayURL&_method=citationSearch&_urlVersion=4&_origin=SDVIALERTASCII&_version=1&_uoikey=B6W6C-4VWB1GF-1&md5=6750958e5c7021a7300a3af4dd4b78af 5) Representation for reciprocal agent?environment interaction Pages 366-376 Tibor Bosse, Catholijn M. Jonker, Jan Treur http://www.sciencedirect.com/science?_ob=GatewayURL&_method=citationSearch&_urlVersion=4&_origin=SDVIALERTASCII&_version=1&_uoikey=B6W6C-4WK48G1-1&md5=a97853a4ae47b488a61f58658f6799b8 Book review 6) Elden, L (2007). Matrix Methods in Data Mining and Pattern Recognition. Philadelphia (USA): Society for Industrial and Applied Mathematics. Pages 377-379 Renato Cordeiro de Amorim http://www.sciencedirect.com/science?_ob=GatewayURL&_method=citationSearch&_urlVersion=4&_origin=SDVIALERTASCII&_version=1&_uoikey=B6W6C-4W3G67G-1&md5=dbc93f38cc06b7297db1a7f585d189d9 =========== See the following Web page for submission, subscription, and other information regarding Cognitive Systems Research: http://www.cogsci.rpi.edu/~rsun/journal.html See http://www.elsevier.com/locate/cogsys for further information regarding accessing these articles. If you have questions about features of ScienceDirect, please access the ScienceDirect Info Site at http://www.info.sciencedirect.com ============ Professor Ron Sun Cognitive Science Department Rensselaer Polytechnic Institute 110 Eighth Street, Carnegie 302A Troy, NY 12180, USA phone: 518-276-3409 fax: 518-276-3017 email: rsun at rpi.edu web: http://www.cogsci.rpi.edu/~rsun ======================================================= From jaakko.peltonen at tkk.fi Mon Oct 5 14:03:48 2009 From: jaakko.peltonen at tkk.fi (jaakko.peltonen@tkk.fi) Date: Mon, 5 Oct 2009 21:03:48 +0300 (EEST) Subject: Connectionists: NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics, Call for contributions Message-ID: ------------------------------------------------------------------ CALL FOR CONTRIBUTIONS NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics Whistler, BC, Canada, December 12, 2009 http://www.dcs.gla.ac.uk/~srogers/lms09/index.htm ------------------------------------------------------------------ Important Dates: ---------------- Submission of extended abstracts: October 27, 2009 Notification of acceptance: November 6, 2009 Workshop Description: --------------------- Learning from multiple sources denotes the problem of jointly learning from a set of (partially) related learning problems / views / tasks. This general concept underlies several subfields receiving increasing interest from the machine learning community, which differ in terms of the assumptions made about the dependency structure between learning problems. In particular, the concept includes topics such as data fusion, transfer learning, multitask learning, multiview learning, and learning under covariate shift. Several approaches for inferring and exploiting complex relationships between data sources have been presented, including both generative and discriminative approaches. The workshop will provide a unified forum for cutting edge research on learning from multiple sources; the workshop will examine the general concept, theory and methods, and will also examine *robotics* as a natural application domain for learning from multiple sources. The workshop will address methodological challenges in the different subtopics and further interaction between them. The intended audience is researchers working in fields of multi-modal learning, data fusion, and robotics. (More detailed background information is available at the workshop website.) The workshop includes a morning session focused on theory/methods, and an afternoon session focused on the robotics application. The workshop is a core event of the PASCAL2 Network of Excellence. PASCAL2 Invited Speakers: ------------------------- Morning Session: Chris Williams - University of Edinburgh Afternoon Session: to be announced Submission Instructions: ------------------------ We invite submission of extended abstracts to the workshop. Extended abstracts should be 2-4 pages, formatted in the NIPS style: http://nips.cc/PaperInformation/StyleFiles Unlike the main NIPS conference, identities of authors do not need to be removed from the extended abstracts. Extended abstracts should be sent in .PDF or .PS file format by email, to either D.Hardoon at cs.ucl.ac.uk or gleen at cis.hut.fi. Acceptance to the workshop will be determined based on peer review of each extended abstract. Submissions are expected to represent high-quality, novel contributions in theory/methods of learning from multiple sources, or high-quality, novel contributions in application of learning from multiple sources to robotics (see below). To encourage participants from the machine learning community to test their algorithms in the domain of robotics, we will make available a dataset, with computed features, representative of open research issues in robotics. Robotics-oriented papers submitted to the workshop are strongly encouraged to contain an experimental evaluation on the database made available by the organizers. The obtained results will be presented by the organizers during the workshop. Submitted extended abstracts may be accepted either as an oral presentation or as a poster presentation; there will be only a limited number of oral presentations in the morning and afternoon sessions. Accepted extended abstracts will be made available online at the workshop website. Depending on the quality of submissions, we will consider preparing a special issue of a journal or a collected volume on the topic of the workshop. A separate call for papers will then be issued after the workshop for the special issue/collected volume. Last year's "Learning from Multiple Sources" workshop led to a special issue in Machine Learning (currently in progress). Organisers ---------- * Barbara Caputo - Idiap Research Institute. * Nicol? Cesa-Bianchi - Universit? degli Studi di Milan. * David Hardoon - Institute for Infocomm Research (I2R). * Gayle Leen - Helsinki University of Technology. * Francesco Orabona - Idiap Research Institure. * Jaakko Peltonen - Helsinki University of Technology. * Simon Rogers - University of Glasgow. Programme Committee ------------------- * Cedric Archambeau - Xerox Research. * Andreas Argyriou - Toyota Technological Institute. * Claudio Gentile - Universit? dell'Insubria. * Mark Girolami - University of Glasgow. * Samuel Kaski - Helsinki University of Technology. * Arto Klami - Helsinki University of Technology. * John Shawe-Taylor - University College London. * Giorgio Valentini - Universit? degli Studi di Milan. Contact Persons --------------- For questions about the workshop, contact David R. Hardoon at D.Hardoon AT cs.ucl.ac.uk. From mpd37 at cam.ac.uk Tue Oct 6 12:01:48 2009 From: mpd37 at cam.ac.uk (Marc Deisenroth) Date: Tue, 06 Oct 2009 17:01:48 +0100 Subject: Connectionists: Probabilistic Approaches for Robotics and Control (NIPS workshop) - 2nd call for contributions Message-ID: <4ACB69EC.70106@cam.ac.uk> ################################################################ CALL FOR CONTRIBUTIONS NIPS workshop on *Probabilistic Approaches for Robotics and Control* (supported by PASCAL 2) ################################################################ *Workshop dates* Friday, December 11, 2009 *Workshop location* Whistler, B.C., Canada, at the Westin Resort and Spa and the Hilton Whistler Resort and Spa *Poster submission* Please send an extended abstract of max. 1 page describing the poster you intend to present to mpd37 at cam.ac.uk Choose a format of your liking, e.g., the standard NIPS template. The *deadline for abstract submissions* is October 17, 2009. The *notification* will be October 26, 2009. *Workshop homepage* http://mlg.eng.cam.ac.uk/marc/nipsWS09 *Conference homepage* http://nips.cc *Workshop Abstract* During the last decade, many areas of Bayesian machine learning have reached a high level of maturity. This has resulted in a variety of theoretically sound and efficient algorithms for learning and inference in the presence of uncertainty. However, in the context of control, robotics, and reinforcement learning, uncertainty has not yet been treated with comparable rigor despite its central role in risk-sensitive control, sensorimotor control, robust control, and cautious control. A consistent treatment of uncertainty is also essential when dealing with stochastic policies, incomplete state information, and exploration strategies. A typical situation where uncertainty comes into play is when the exact state transition dynamics are unknown and only limited or no expert knowledge is available and/or affordable. One option is to learn a model from data. However, if the model is too far off, this approach can result in arbitrarily bad solutions. This model bias can be sidestepped by the use of flexible model-free methods. The disadvantage of model-free methods is that they do not generalize and often make less efficient use of data. Therefore, they often need more trials than feasible to solve a problem on a real-world system. A probabilistic model could be used for efficient use of data while alleviating model bias by explicitly representing and incorporating uncertainty. The use of probabilistic approaches requires (approximate) inference algorithms, where Bayesian machine learning can come into play. Although probabilistic modeling and inference conceptually fit into this context, they are not widespread in robotics, control, and reinforcement learning. Hence, this workshop aims to bring researchers together to discuss the need, the theoretical properties, and the practical implications of probabilistic methods in control, robotics, and reinforcement learning. One particular focus will be on probabilistic reinforcement learning approaches that profit recent developments in optimal control which show that the problem can be substantially simplified if certain structure is imposed. The simplifications include linearity of the (Hamilton-Jacobi) Bellman equation. The duality with Bayesian estimation allow for analytical computation of the optimal control laws and closed form expressions of the optimal value functions. Format The workshop will consist of short invited presentations and a session with contributed posters (plus poster spotlight). Topics (from a theoretical and practical perspective) to be addressed include, but are not limited to: - How can we efficiently plan and act in the presence of uncertainty in states/rewards/observations/environment? - Shall we model the lack of knowledge or can we simply ignore it? - How can prior knowledge (e.g., expert knowledge and domain knowledge) be incorporated? - How much manual tuning and human insight (e.g., domain knowledge) is a) required and b) available to achieve good performance? - Is there a principled way to account for imprecise models and model bias? - What roles should probabilistic models play in control? Are they needed at all? - What kinds of probabilistic models are useful? - In traditional control, hand-crafted control laws often prevail since optimal control laws are mostly too aggressive due to model errors while robust control laws can be too conservative since they always assume the worst case. Can "probabilistic control" bridge the gap between robust and optimal control laws? - How can we exploit the linearity of the (Hamilton-Jacobi) Bellman equation and the duality with Bayesian estimation? - Can we compute the optimal control law analytically and is there a closed-form expression of the value function? - How can existing machine learning methods be applied to efficiently solve stochastic control problems? *Invited speakers* Dieter Fox (University of Washington), confirmed Drew Bagnell (CMU), pending Evangelos Theodorou (USC), confirmed Jovan Popovic (MIT), confirmed Konrad Koerding (Northwestern University), confirmed Marc Toussaint (TU Berlin), confirmed Miroslav Karny (Academy of Sciences of the Czech Republic), confirmed Roderick Murray-Smith (University of Glasgow), confirmed Bert Kappen (University of Nijmegen), confirmed Emanuel Todorov (University of Washington), confirmed *Support* The workshop is supported by PASCAL2 and the Technical Committee on Robot Learning *Organizers* Marc Peter Deisenroth Bert Kappen Emanuel Todorov Duy Nguyen-Tuong Carl Edward Rasmussen Jan Peters From hiro at brain.riken.jp Wed Oct 7 01:01:36 2009 From: hiro at brain.riken.jp (hiroyuki nakahara) Date: Wed, 07 Oct 2009 14:01:36 +0900 Subject: Connectionists: Postdoctoral Positions in Theoretical Neuroscience, at RIKEN BSI Message-ID: <20091007135714.6683.HIRO@brain.riken.jp> Postdoctoral positions available Postdoctoral positions are available for studying neural valuation and associated decision making in the laboratory of Dr. Hiro Nakahara (the Laboratory for Integrated Theoretical Neuroscience) at RIKEN Brain Science Institute. A major focus of the research for these positions is to develop theoretical and/or computational models for a normative understanding (e.g. reinforcement learning framework and statistical inference) of neural valuation/decision systems, further linking to a mechanistic or circuit-level understanding, in consideration of the rich repertory of behavioral and neurophysiological data. For general information on our laboratory, see http://www.itn.brain.riken.jp. Applicants should have a Ph.D. We seek exceptionally talented candidates with a strong background in theoretical neuroscience. Solid experience in theoretical neuroscience research, with a strong analytical background and a ready ability to acquire new information in experimental literature (e.g. cognitive and behavioral neuroscience), is essential. Research experience in decision making and/or valuation, possibly including basal ganglia functions (and/or the frontal cortices), is a major plus. Proficiency in computer programming (e.g. Matlab or equivalent) is expected. Good communication and writing skills are essential. A good balance of independence and collegiality in research is required. The RIKEN Brain Science Institute (URL: www.brain.riken.jp) is located near Tokyo, Japan. It uses English as the working language, and provides an international, vigorous, and interactive environment not only for computational neuroscience but also for a broad range of disciplines in neuroscience. These positions can start immediately, but consideration will be also given to candidates who would prefer a later start date. Consideration of applications will start immediately, but all applications sent to us by November 10th will be equally considered. Appointment is on an annual basis, and starting salaries will be commensurate with relevant ability and experience. Subsequent contracts will be determined and renewed annually, upon review, for possibly up to five years. Interested candidates should apply to itninfo at brain.riken.jp with the following: a cover letter, CV, a statement of research skills and interests, contact details of three references with a brief description of your relationship to each reference, and (optional) any additional information you think might be useful (e.g. additional skills and background, general interests, and so on). Informal enquiries are also welcome. Hiro Nakahara Lab for Integrated Theoretical Neuroscience RIKEN Brain Science Institute 2-1 Hirosawa Wako Saitama, 351-0198 Japan http://www.itn.brain.riken.jp -- hiroyuki nakahara http://www.itn.brain.riken.jp From kirsch at bccn.uni-freiburg.de Wed Oct 7 04:26:36 2009 From: kirsch at bccn.uni-freiburg.de (Janina Kirsch) Date: Wed, 7 Oct 2009 10:26:36 +0200 Subject: Connectionists: PhD-Positions in Neuroinformatics, including Computational Neuroscience (Erasmus Mundus Program) Message-ID: The Erasmus Mundus Joint Doctoral Program "EuroSPIN " (European Study Programme in Neuroinformatics) is inviting applications from students having a solid background in mathematics, physics, computer sciences, biochemistry or neuroscience (on a master level or equivalent), in all cases with computer science skills. Documented interest in research like activities (e.g. demonstrated in the form of master thesis work, or participation in research related activities) is of large importance. Also fluency in English is requested. Neuroinformatics combines neuroscience and informatics research to develop and apply computational tools and approaches that are essential for understanding the structure and function of the brain. Four partners participate in EuroSPIN: - KTH Royal Institute of Technology, Sweden - University of Edinburg (UoE), UK - National Centre for Biological Science (NCBS), India - Albert-Ludwigs-Universit?t Freiburg (ALUF), Germany These four partners are all research leaders in the Neuroinformatics field, but they have complementary strengths. In addition, two associated partners, the Honda Research Institute and Nordita, participate. Each student will spend most of the time at two of the partner universities, and also receive a joint (or double) PhD degree following a successful completion of the studies. The mobility periods, as well as the courses a student will follow, are tailored individually based on: a) the PhD students background; b) which constellations of partners that are involved, as well as c) the specific research project. During the PhD period each student has one main supervisor from each of the two universities that grant the PhD degree. During the application process, the students are asked to indicate their preferences with regard to constellations of partners, and also preferred project ideas/areas can be indicated and motivated. There are excellent scholarship opportunities for students accepted to an Erasmus Mundus Joint Doctorate programme. A stipend or employment contract will be given to all selected PhD students during the study time, which is between 3-4 years. If you are interested, go to our homepage: http://www.kth.se/studies/phd/eurospin?l=en_UK. Deadline for Application (non-EU students): December 15, 2009. Deadline for Application (EU students): about March 2009 (will be announced). -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091007/3b574366/attachment-0001.html From ncpw12 at googlemail.com Wed Oct 7 05:19:52 2009 From: ncpw12 at googlemail.com (NCPW12) Date: Wed, 7 Oct 2009 10:19:52 +0100 Subject: Connectionists: NCPW12 Call for Abstracts In-Reply-To: <63e97b060910051032p6ac2775dje438cbaa95b1eec0@mail.gmail.com> References: <63e97b060910051032p6ac2775dje438cbaa95b1eec0@mail.gmail.com> Message-ID: <63e97b060910070219t462b6122p1922f752e4ac58a6@mail.gmail.com> ***Apologies for cross-postings*** *** CALL FOR ABSTRACTS *** NCPW12 12th Neural Computation and Psychology Workshop London, UK 8-10 April 2010 http://www.bbk.ac.uk/psyc/staff/academic/eddyjdavelaar/ncpw12/index.php ********************** We cordially invite you to participate in the 12th Neural Computation and Psychology Workshop (NCPW12), to be held at Birkbeck, University of London, from Thursday 8th to Saturday 10th April 2010. This well-established and lively workshop aims at bringing together researchers from different disciplines such as artificial intelligence, cognitive science, computer science, neurobiology, philosophy and psychology to discuss their work on models of cognitive processes. Previous themes have encompassed categorisation, language, memory, development, action. This year?s theme is From Theory to Applications, and papers must be about emergent models -- frequently, but not necessarily -- of the connectionist/neural network kind, applied to cognition. These workshops have always been characterised by their limited size, high quality papers, the absence of parallel talk sessions, and a schedule that is explicitly designed to encourage interaction among the researchers present in an informal setting. ********************** DEADLINE FOR SUBMISSION OF ABSTRACTS: December 18th, 2009 Abstract submission is now open. For more information see the conference website at: http://www.bbk.ac.uk/psyc/staff/academic/eddyjdavelaar/ncpw12/index.php Notification of acceptance: Approx. 11th January 2010. ********************** Dates & Deadlines Abstract submission: 18 December 2009 Notification of acceptance: 11 January 2010 Conference: 8-10 April 2010 = = = = = = = = = = = = = = = = = = = = = = = = = = = = = = Dr Eddy J. Davelaar Department of Psychological Sciences Birkbeck, University of London, WC1E 7HX http://www.bbk.ac.uk/psyc/staff/academic/eddyjdavelaar -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091007/74f44d18/attachment.html From gluck at pavlov.rutgers.edu Wed Oct 7 06:03:40 2009 From: gluck at pavlov.rutgers.edu (Mark A. Gluck) Date: Wed, 7 Oct 2009 06:03:40 -0400 Subject: Connectionists: Postdoc at Rutgers, NJ: Cognitive/Computational Neuropsych of Learning and Decision Making Message-ID: COGNITIVE / COMPUTATIONAL NEUROPSYCHOLOGY OF LEARNING & DECISION MAKING A Postdoctoral Fellowship Position ___ NOTE: Applicants must be US Citizens or green card holders. ____ We seek a postdoctoral applicant for a position at Rutgers University - Newark in New Jersey to work as part of an international and interdisciplinary team studying the cognitive and computational neuropsychology of learning and decision making. We seek an applicant to work in either or both of the following capacities: (1) Conducting clinical neuropsychological behavioral studies of patients with Parkinson's disease, drug addiction, schizophrenia and depression. We study how these disorders impact learning and decision making, how individual genetic differences (especially those relating to serotonin and dopamine) impact these abilitites, and how current pharmacological treatments for these disorders impact cognition. In addition, we seek to understand how individual differences in cognitive functioning among patients can help identify clinically-relevant subtypes of these disorders, especially with regard to linking cognitive and psychiatric side effects. Some of these studies are done in parallel with analogous functional brain imaging or animal studies. Collaborative projects include local partners in NY and NJ, as well as international collaborations in Italy, Turkey, Hungary, Israel, Egypt, and the Palestinian West Bank. (2). Implementing and testing neurocomputational "neural-network" models of the role of fronto-striatal and medial temporal lobe function in learning and decision making, and how these are impacted by neurological/psychiatric disorders and pharmacological manipulations. Modeling is done at a systems/cognitive level (not molecular/cellular level). Substantial prior experience with either (1) human neuropsychological studies of cognition or (2) neurocomputational modeling is required. We are located in downtown Newark, 13 miles outside of New York City. Applicants should see http://www.gluck.edu for more information on our lab and lab personnel and to familiarize yourself with recent research and publications (all available on line as PDFs). To apply, please send a cover letter that summarizes your relevant experience and how you see your skills and interets fitting into the above research programs. Include also your CV, summary of your past and future research interests, and a list of 3 potential references, and send as a single email with attachments to: Dr. Mark A. Gluck at gluck at pavlov.rutgers.edu -- ___________________________________________ Dr. Mark A. Gluck, Professor Co-Director, Rutgers Memory Disorders Project Center for Molecular and Behavioral Neuroscience Rutgers University Phone: (973) 353-3668/3298 197 University Ave. Fax: (973) 353-1272 Newark, New Jersey 07102 Email: gluck at pavlov.rutgers.edu Lab: http://www.gluck.edu Public Information: http://www.memory.rutgers.edu Memory Loss & Brain Newsletter: http://www.memorylossonline.com -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091007/ce8b7704/attachment.html From gluck at pavlov.rutgers.edu Wed Oct 7 06:03:17 2009 From: gluck at pavlov.rutgers.edu (Mark A. Gluck) Date: Wed, 7 Oct 2009 06:03:17 -0400 Subject: Connectionists: New paper: Neurocomputational model of tonic and phasic dopamine in cognitive deficits in Parkinson's Message-ID: Dear Colleagues, A PDF of our new neurocomputational modeling paper: Guthrie, M., Myers, C. E., & Gluck, M. A. (2009/In press). A neurocomputational model of tonic and phasic dopamine in action selection: A comparison with cognitive deficits in Parkinson's disease. Behavioral Brain Research. has been posted online to our lab web site at: http://www.gluck.edu/pdf/BBR_5811.pdf [if you have trouble with this URL, it is also linked from our home page at http://www.gluck.edu/ ] An ABSTRACT for this paper is as follows: The striatal dopamine signal has multiple facets; tonic level, phasic rise and fall, and variation of the phasic rise/fall depending on the expectation of reward/punishment.We have developed a network model of the striatal direct pathway using an ionic current level model of the medium spiny neuron that incorporates currents sensitive to changes in the tonic level of dopamine. The model neurons in the network learn action selection based on a novel set of mathematical rules that incorporate the phasic change in the dopamine signal. This network model is capable of learning to perform a sequence learning task that in humans is thought to be dependent on the basal ganglia. When both tonic and phasic levels of dopamine are decreased, as would be expected in unmedicated Parkinson's disease (PD), the model reproduces the deficits seen in a human PD group off medication. When the tonic level is increased to normal, but with reduced phasic increases and decreases in response to reward and punishment, respectively, as would be expected in PD medicated with L-Dopa, the model again reproduces the human data. These findings support the view that the cognitive dysfunctions seen in Parkinson's disease are not solely either due to the decreased tonic level of dopamine or to the decreased responsiveness of the phasic dopamine signal to reward and punishment, but to a combination of the two factors that varies dependent on disease stage and medication status. As always, we welcome and appreciate comments, feedback, and pointers to relevant published or unpublished research. This paper is one in a series of forthcoming theoretical papers that seek to better understand the nature of the cognitive deficits in Parkison's disease through modeling frontal and/or striatal dopamine function in learning and decision making. Regards, Mark Gluck ___________________________________ Dr. Mark A. Gluck, Professor Co-Director, Rutgers Memory Disorders Project Center for Molecular and Behavioral Neuroscience Rutgers University 197 University Ave. Newark, New Jersey 07102 Web: http://www.gluck.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091007/17213bda/attachment-0001.html From wsenn at cns.unibe.ch Wed Oct 7 11:39:50 2009 From: wsenn at cns.unibe.ch (Walter Senn) Date: Wed, 07 Oct 2009 17:39:50 +0200 Subject: Connectionists: Biological Cybernetics: vol 101, issue 3 -- Table of Content Message-ID: <4ACCB646.4020502@cns.unibe.ch> Biological Cybernetics: vol 101, issue 3 --- Table of Content "What can be learnt from analysing insect orientation flights using probabilistic SLAM?" Bartholomew Baddeley, Andrew Philippides, Paul Graham, Natalie Hempel de Ibarra, Thomas Collett & Phillip Husbands Page 169 - 182 http://www.springerlink.com/content/22170w046237m86q/ "An investigation of the evolutionary origin of reciprocal communication using simulated autonomous agents" Elio Tuci Page 183 - 199 http://www.springerlink.com/content/y03w6258364v111k/ "A phase dynamic model of systematic error in simple copying tasks" Saguna Dubey, Sandeep Sambaraju, Sarat Chandra Cautha, Vednath Arya & V. S. Chakravarthy Page 201 - 213 http://www.springerlink.com/content/p78u0443j98003p7/ "Object orientation in two dimensional grasp with friction towards minimization of gripping power" Satoshi Ito, Shouta Takeuchi & Minoru Sasaki Page 215 - 226 http://www.springerlink.com/content/82416w433001422r/ "Spike-timing-dependent plasticity leads to gamma band responses in a neural networks" Ingo Fr?nd, Frank W. Ohl & Christoph S. Herrmann Page 227 - 240 http://www.springerlink.com/content/w35205g33550776j/ "Washout filter aided mean field feedback desynchronization in an ensemble of globally coupled neural oscillators" Ming Luo, Yongjun Wu & Jianhua Peng Page 241 - 246 http://www.springerlink.com/content/12651284085x1642/ Biological Cybernetics, all issues: http://www.springerlink.com/content/100465/ From n.berthouze at ucl.ac.uk Wed Oct 7 16:21:53 2009 From: n.berthouze at ucl.ac.uk (Nadia Berthouze) Date: Wed, 07 Oct 2009 21:21:53 +0100 Subject: Connectionists: CALL FOR PARTICIPATION - EPIROB'09 Message-ID: <4ACCF861.6050107@ucl.ac.uk> CALL FOR PARTICIPATION - EPIROB'09 9th International Conference on Epigenetic Robotics ? Modeling Cognitive Development in Robotic Systems Venice, Italy, 12th-14th november, Centro Culturale Don Orione Artigianelli http://www.epigenetic-robotics.org/ Registration is now open! === Keynote speakers: Kim Bard, University of Portsmouth, UK Cynthia Breazeal, MIT Media Lab, US Heidi Keller, University of Osnabr?ck, Germany David Leavens, Sussex University, UK === Topics Epigenetic robotics includes the two-fold goal of understanding biological systems by the interdisciplinary integration between social/life and engineering sciences and, simultaneously, that of enabling robots and other artificial systems to autonomously develop skills for any particular environment (instead of programming them to solve particular goals for a specific environment). Interdisciplinary theory and empirical evidence are used to inform epigenetic robotic models, and these models can be used as theoretical tools to make experimental predictions in developmental psychology and other disciplines studying cognitive development in living systems. SPECIAL TOPIC of EPIROB'09: This year's edition of Epigenetic Robotics will have a special focus on emotional and social development, particularly addressed by keynote speakers and special discussion and working groups. However, submissions are welcome regarding all aspects of the study of cognitive development. === 14th november afternoon: Joint event with the International Workshop on Intrinsically Motivated Robots, organized by the EU project IM-CLeVeR on November 14-17, 2009, Venice, at the same same than Epirob. http://im-clever.noze.it/announcements/events/im-clever-international-workshop Topic: Intrinsic Motivation and Socio-Emotional Development ==Keynote speakers: Jacqueline Nadel, CNRS, Universit? Pierre-et-Maris Curie/La Salp?tri?re, France Pierre-Yves Oudeyer, INRIA, France The workshop will start in the afternoon of November 14 with a joint session co-organized by EpiRob'09, FEELIX GROWING and IM-CLeVeR. This workshop will be free of charge for EpiRob'09 delegates, within the limits of space available. === Organizing committee General and program chair: Lola Canamero, Univ. Hertfordshire, UK Program co-chair: Pierre-Yves Oudeyer, INRIA, France Publicity co-chairs: Hideki Kozima, Miyagi Univ., Japan Nadia Bianchi-Berthouze, Univ. College London, UK Aude Billard, EPFL, Switzerland Publication chair: Christian Balkenius, Lund University, Sweden We are looking forward to seeing you in Venice! From terry at salk.edu Wed Oct 7 21:48:25 2009 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 07 Oct 2009 18:48:25 -0700 Subject: Connectionists: NEURAL COMPUTATION - November, 2009 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 21, Number 11 - November 1, 2009 LETTERS On the Maximization of Information Flow Between Spiking Neurons Lucas Parra, Jeffrey Beck, and Anthony Bell Cortical Circuitry Implementing Graphical Models Shai Litvak and Shimon Ullman Electrical Coupling Promotes Fidelity of Responses in the Networks of Model Neurons Georgi S. Medvedev Maximum Likelihood Decoding of Neuronal Inputs from an Interspike Interval Distribution Xuejuan Zhang, Gongqiang You, Tianping Chen, and Jianfeng Feng Classifcation of Correlated Patterns with a Configurable Analog VLSI Neural Network of Spiking Neurons and Self-Regulating Plastic Synapses Massimilian Giulioni, MMario Pannunzi, Davide Badoni, Vittorio Dante, and Paolo Del Giudice Approximate Learning Algorithm in Boltzmann Machines Muneki Yasuda and Kazuyuki Tanaka Robust Kernel Principal Component Analysis Su-Yun Huang, Yi-Ren Yeh, and Shinto Eguchi Adaptive Capability of Recurrent Neural Networks with Fixed Weights for Series-Parallel System Identifcation James Ting-Ho Lo Partial Orders of Similarity Differences Invariant between EEG-Recorded Brain and Perceptual Representations of Language Patrick Suppes, Marc Perreau-Guimaraes, and Dik Kin Wong ----- ON-LINE - http://www.mitpressjournals.org/loi/neco SUBSCRIPTIONS - 2009 - VOLUME 21 - 12 ISSUES USA/Canada Others Electronic only Student/Retired $60 $123 $54 Individual $110 $173 $99 Institution $849 $912 $756 MIT Press Journals, 238 Main Street, Suite 500, Cambridge, MA 02142-9902. Tel: (617) 253-2889 FAX: (617) 577-1545 journals-orders at mit.edu http://mitpressjournals.org/neuralcomp ----- From ndg at MIT.EDU Fri Oct 9 09:23:59 2009 From: ndg at MIT.EDU (Noah Goodman) Date: Fri, 9 Oct 2009 09:23:59 -0400 Subject: Connectionists: CALL FOR CONTRIBUTIONS: NIPS 2009 Workshop, "Bounded-rational analyses of human cognition" Message-ID: <9542A04D-7521-45C9-A452-EBE14B263F06@mit.edu> ---------------------------------- CALL FOR CONTRIBUTIONS NIPS 2009 Workshop: Bounded-rational analyses of human cognition: Bayesian models, approximate inference, and the brain. http://www.mit.edu/~ndg/NIPS09Workshop.html Whistler, BC, Canada. Dec 12, 2009. ---------------------------------- We invite poster submissions for the NIPS 2009 workshop "Bounded- rational analyses of human cognition: Bayesian models, approximate inference, and the brain". Relevant topics include (but are not limited to): state-of-the-art algorithms for bounded and time-limited inference, process-level limitations on human Bayesian inference, inference algorithms in humans, and neural implementations of Bayesian inference algorithms. Abstracts, no longer than one page, may be submitted by email to: ndg at mit.edu , no later than October 31, 2009. Please include "NIPS Workshop Abstract" in the subject of your email. DESCRIPTION Bayesian, or "rational", accounts of human cognition have enjoyed much success in recent years: human behavior is well described by probabilistic inference in low-level perceptual and motor tasks as well as high level cognitive tasks like category and concept learning, language, and theory of mind. However, these models are typically defined at the abstract "computational" level: they successfully describe the computational task solved by human cognition without committing to the algorithm which carries it out. Bayesian models usually assume unbounded cognitive resources available for computation, yet traditional cognitive psychology has emphasized the severe limitations of human cognition. Thus, a key challenge for the Bayesian approach to cognition is to describe the algorithms used to cary out approximate probabilistic inference using the bounded computational resources of the human brain. Inspired by the success of Monte Carlo methods in machine learning, several different groups have suggested that humans make inferences not by manipulating whole distributions, but my drawing a small number of samples from the appropriate posterior distribution. Monte Carlo algorithms are attractive as algorithmic models of cognition both because of they have been used to do inference in a wide variety of structured probabilistic models, scaling to complex situations while minimizing the curse of dimensionality, and because they use resources efficiently, and degrade gracefully when time does not permit many samples to be generated. Indeed, given parsimonious assumptions about the cost of obtaining a sample for a bounded agent, it is often best to make decisions using just one sample. The claim that human cognition works by sampling identifies the broad class of Monte Carlo algorithms as candidate cognitive process models. Recent evidence from human behavior supports this coarse description of human inference: people seem to operate with a limited set of samples at a time. Further narrowing the class of algorithm makes additional predictions if the samples drawn by these algorithms are imperfect samples (not exact samples from the posterior distribution). That is, while most Monte Carlo algorithms yield unbiased estimators given unlimited resources, they all have characteristic biases and dynamics in practice -- it is these biases and dynamics which result in process-level predictions about human cognition. For instance, it has been argued that the characteristic order effects exhibited by sequential Monte Carlo algorithms (particle filters) when run with few particles can explain the primacy and recency effects observed in human category learning, and the "garden path" phenomena of human sentence processing. Similarly, others have argued that the temporal correlation of samples obtained from Markov Chain Monte Carlo (MCMC) sampling can account for bistable percepts in visual processing. Ultimately the processes of human cognition must be implemented in the brain. Relatively little work has examined how probabilistic inference may be carried out by neural mechanisms, and even less of this work has been based on Monte Carlo algorithms. Several different neural implementations of probabilistic inference, both approximate and exact, have been proposed, but the relationship among these implementations and to algorithmic and behavioral constraints remains to be understood. Accordingly, this workshop will foster discussion of neural implementations in light of work on bounded-rational cognitive processes. The goal of this workshop is to explore the connections between Bayesian models of cognition, human cognitive processes, modern inference algorithms, and neural information processing. We believe that this will be an exciting opportunity to make progress on a set of interlocking questions: Can we derive precise predictions about the dynamics of human cognition from state-of-the-art inference algorithms? Can machine learning be improved by understanding the efficiency tradeoffs made by human cognition? Can descriptions of neural behavior be constrained by theories of human inference processes? ORGANIZERS: Noah Goodman Ed Vul Tom Griffiths Josh Tenenbaum INVITED SPEAKERS (confirmed) Matt Botvinik Noah Goodman Tom Griffiths Stuart Russell Paul Schrater Ed Vul Jerry Zhu WORKSHOP FORMAT 8:00 introductory remarks 8:10 1st talk 8:40 2nd talk 9:10 break 9:30 3rd talk 10:00 4th talk 10:30 discussion 11:00 - 1:00 posters 4:00 5th talk 4:30 6th talk 5:00 7th talk 5:30 8th talk 6:00 discussion -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091009/3cd080ca/attachment-0001.html From jch1 at cs.cmu.edu Fri Oct 9 12:46:14 2009 From: jch1 at cs.cmu.edu (Jonathan Huang) Date: Fri, 9 Oct 2009 12:46:14 -0400 (EDT) Subject: Connectionists: CfP: NIPS 2009 Workshop on Learning with Orderings Message-ID: <50652.128.2.211.60.1255106774.squirrel@webmail.cs.cmu.edu> ------------------------------ -------------------------------------------- LEARNING WITH ORDERINGS workshop at NIPS'09 Saturday, December 12, Whistler, B.C. http://www.select.cs.cmu.edu/meetings/nips09perm/ -------------------------------------------------------------------------- The scope of the workshop is everything in ML related to permutations, in particular: Ranking and search Preference elicitation Social choice and voting theory Tracking and identity management Natural language processing Structure learning for Bayesian networks Multi-way classification and other categorization tasks Graph matching Partner workshop: "Advances in Ranking" http://web.mit.edu/shivani/www/Ranking-NIPS-09/ Organizers: Tiberio Caetano, NICTA Carlos Guestrin, CMU Jonathan Huang, CMU Guy Lebanon, Georgia Tech Risi Kondor, Caltech Marina Meila, UW Confirmed speakers: Persi Diaconis, Stanford Manfred Warmuth, UCSC Xiaoye Jiang, Stanford Regina Barzilay, MIT Risi Kondor, Caltech Dan Rockmore, Dartmouth College Brendan Murphy, University College Dublin Call for abstracts: The workshop will include a combined poster/demo session. We invite participants to submit 2-page extended abstracts (NIPS format) to LearningWithOrderings at gmail.com describing the technical content of proposed posters and computer demos. The submission deadline is November 9. The review process will be coordinated with the "Advances in Ranking" workshop. Please do not submit the same work to both workshops. From murphyk at cs.ubc.ca Fri Oct 9 12:44:21 2009 From: murphyk at cs.ubc.ca (Kevin Murphy) Date: Fri, 9 Oct 2009 09:44:21 -0700 (PDT) Subject: Connectionists: postdoc opportunity at UBC in vision/learning Message-ID: I am pleased to announce that the Canadian Institute For Advanced Research (CIFAR) is funding a 2 year Junior Fellow position as part of its Neural Computation and Adaptive Perception (NCAP) program. The Fellow will be based at the University of British Columbia (UBC) in Vancouver, and will work with one or more of Kevin Murphy, Nando de Freitas, and David Lowe. The starting date is flexible, but it is expected to be sometime between June and September 2010. The precise topics of research are still to be determined, but will be related to using unsupervised, semi-supervised and 'deep learning' to solve interesting vision problems, such as generic object detection and localisation, image segmentation and labeling, video parsing, feature discovery, active vision, etc. The ideal candidate will have a strong research track record in learning and vision, and an interest in neuroscience, as well as excellent communication skills, and a strong potential to collaborate with other NCAP program members. The candidate will be expected to fully participate in NCAP meetings (usually 2-3 per year, typically in Vancouver or Toronto), and CIFARs Junior Fellow Academy, interacting with peers in CIFAR's other research programs. Please send a 2 page research statement and a recent CV to murphyk at cs.ubc.ca by 15 Jan 2010. Also, please arrange to have 3 letters of reference sent to the same address, by 15 Jan 2010. At least one of these letters should be 'arms length' and from a senior figure in the field of vision and/or learning and/or computational neuroscience. More details on the NCAP program can be found here: http://www2.cifar.ca/research/neural-computation-and-adaptive-perception-program/ Please read this page before applying! Kevin Murphy From marcus.hutter at gmx.net Fri Oct 9 05:07:18 2009 From: marcus.hutter at gmx.net (Marcus Hutter) Date: Fri, 9 Oct 2009 20:07:18 +1100 Subject: Connectionists: AGI-10 Submission Deadline Extension Message-ID: <0ed201ca48c4$2b48c660$c09412c6@crl174ml1s> Hi All, Upon popular request, the deadline for submitting papers to the Third Conference on Artificial General Intelligence, AGI-10 (in Lugano, Switzerland, March 5-8 2010) has been extended to November 1. Continuing the mission of the highly successful First and Second AGI Conferences, AGI-10 will gather an international group of leading academic and industry researchers involved in serious scientific and engineering work aimed directly toward the goal of artificial general intelligence. This is the only major conference series devoted wholly and specifically to the creation of AI systems possessing general intelligence at the human level and ultimately beyond. By gathering together active researchers in the field, for presentation of results and discussion of ideas, we accelerate our progress toward our common goal. This is the first AGI conference to be held outside the US and we are looking forward to a lively mixture of perspectives. Lugano is a beautiful locale and we are sure it will be an inspiring place to collaborate on creating the future of AGI! The Call for Papers is here http://agiri.org/AGI_10_Poster.pdf http://agi-conf.org/2010/call-for-papers/ and the conference website is here http://agi-conf.org/2010 With Rich Sutton as keynote speaker, AGI-10 is shaping up to be even better than its predecessors. We look forward to receiving your papers Thanks, Ben Goertzel, Novamente LLC. AGI Conference Series Chair Marcus Hutter, Australian National University, AGI-10 Conference Chair Itamar Arel, University of Tennessee, AGI-10 Program Committee Co-chair Eric Baum, Baum Research Enterprises, AGI-10 Program Committee Co-chair Juergen Schmidhuber, IDSIA, Lugano, Local Conference Chair From raphael.ritz at incf.org Fri Oct 9 05:19:29 2009 From: raphael.ritz at incf.org (Raphael Ritz) Date: Fri, 09 Oct 2009 11:19:29 +0200 Subject: Connectionists: INCF at SfN 2009, Chicago Message-ID: <4ACF0021.7070805@incf.org> Dear colleagues, like in previous years, the International Neuroinformatics Coordinating Facility (INCF; www.incf.org) is presenting selected neuroinformatics projects at the Annual Meeting of the Society for Neuroscience (booth #2100 in the Exhibit Hall of the Chicago Convention Center, Sunday, October 18 - Wednesday, October 21). See below for this years program or download the flyer from http://incf.org/documents/news-material/neuroscience-2009/Demoprogram-SfN2009.pdf In addition, we would like to invite you all to the social event "Neuroscience 2.0 - Networking data, tools, and people", held on Monday October 19, 6:30 p.m. - 9:00 p.m. in Lake Erie Room at the Chicago Hilton. And on Tuesday, Oct. 20 6:45?8:45 p.m. there will be the "Computational Neuroscience Social: Not an Oxymoron!" with brief presentations at McCormick Place: Room N228. Last but not least I would like to point you to the "Related Abstract Search Tool" - a contribution from the INCF Japan Node to help you organize your SfN 2009 meeting attendance: http://ras.ni.brain.riken.jp/SfN2009/ See you in Chicago? Raphael ============================================================================= Schedule of the Neuroinformatics demonstrations at the INCF Booth at SfN 2009 ============================================================================= Location: booth #2100 in the Exhibit Hall of the Chicago Convention Center Sunday morning, October 18, 2009, 9:30am-12:30pm ------------------------------------------------ - Reaching 500 models in ModelDB: implications for advances in neuronal integration Morse TM, Carnevale NT, Migliore M, Hines ML, and Shepherd GM - Modelling Large-Scale Neuronal Networks with the NEST Topology Module Plesser HE, Austvoll K, Diesmann M, Eppler JM, Gewaltig M-O, and Morrison A Sunday afternoon, October 18, 2009, 1:30pm-4:30pm ------------------------------------------------- - Spectral Analysis of Neural Time Series Data: An introduction to Chronux Bokil H and Mitra P - BrainInfo Online 3D Macaque Brain Atlas Dubach MF and Bowden DM ========================================================= Monday morning, October 19, 2009, 9:30am-12:30pm ------------------------------------------------ - INCF Japan Node (J-Node) and neuroinformatics platforms: Integrative Brain Research Platform, Cerebellar Development Transcriptome Database Platform Usui S, Takao K, Wagatsuma H, and Kamiji NL - The INCF Digital Brain Atlasing Program: Community Built Infrastructure Spanning Multiple Atlas Spaces Hawrylycz M, Lau C, Larson S, and Boline J Monday afternoon, October 19, 2009, 1:30pm-4:30pm ------------------------------------------------- - INCF Japan Node (J-Node) and neuroinformatics platforms: Dynamic Brain Platform, Related Abstract Search Tool Usui S, Wagatsuma H, Asai Y, Inagaki K, and Kamiji NL - The CARMEN Portal for Sharing and Analysis of Neurophysiological Data Ingram C, Smith L, Simonotto J, Williams L, Hiden H, Sernagor E, Jackson T, and Austin J ========================================================= Tuesday morning, October 20, 2009, 9:30am-12:30pm ------------------------------------------------- - Neuroinformatics resources for computational neuroanatomy Polavaram S, Hamilton D, and Ascoli GA - The Rodent Brain Workbench: Web-enabled Brain Mapping at Microscopic Resolution Bjaalie JG, Leergaard TB, Kj?nigsen LJ, Zakiewicz I, van Dongen YC, Odeh F, Papp EA, Ramachandran M, Darine DA, and Moene IA Tuesday afternoon, October 20, 2009, 1:30pm-4:30pm -------------------------------------------------- - Modelling Network Diseases: From Retinal Dysfunction to Epilepsy Simonotto J, Marcelino J, and Kaiser M - Scalable Brain Atlas Viewer: NeuroLex concepts in interactive 3D-context Bakker R, Larson S, Bezgin G, Heeren D, and K?tter R ========================================================= Wednesday morning, October 21, 2009, 9:30am-12:30pm --------------------------------------------------- - NeuroLex.org - a semantic wiki for neuroinformatics based on the NIF Standard Ontology Larson S, Maynard S, Imam F, and Martone M - Data Sharing Between NITRC and the INCF Software Center Haselgrove C, Larsson A, Bjaalie JG, Breeze J, Buccigrossi R, Kennedy D, Preuss N, and Ritz R Wednesday afternoon, October 21, 2009, 1:30pm-4:30pm ---------------------------------------------------- - Open forum discussion: Anyone is welcome to give a spontaneous demonstration -- Dr Raphael Ritz Scientific Officer International Neuroinformatics Coordinating Facility Secretariat Karolinska Institutet Nobels v?g 15A SE-171 77 Stockholm Sweden Email: raphael.ritz at incf.org Phone: +46 8 524 87017 Fax: +46 8 524 87150 web: www.incf.org From terry at salk.edu Fri Oct 9 23:55:49 2009 From: terry at salk.edu (Terry Sejnowski) Date: Fri, 09 Oct 2009 20:55:49 -0700 Subject: Connectionists: NIPS 2009 Workshop - "The Curse of Dimensionality Problem: How Does the Brain Solve It?" In-Reply-To: Message-ID: NIPS 2009 Workshop - Whistler Canada - http://nips.cc/Conferences/2009/Program/ Friday, December 11, 2009 7.30 am - 10.30 am 16.00 am - 19.00 pm "The Curse of Dimensionality Problem: How Does the Brain Solve It?" Organizers: Simon Haykin, McMaster University Terry Sejnowski, Salk Institute and UCSD Steven Zucker, Yale University One Day Format: Tutorial Lectures (Haykin, Sejnowski, Zucker) Invited Speakers (Valiant, Tsotsos, Bialek, Tishby, Powell) Abstract: The notion of "Curse of Dimensionality" was coined by Richard Bellman (1961). It refers to the exponential increase in computing a task of interest when extra dimensions are added to an associated mathematical space. For example, it arises in solving dynamic programming and optimal control problems when the dimension of the state vector is large. It also arises in solving learning problems when a finite number of data samples is used to learn a "state of nature, the distribution of which is infinitely large." Much has been written on the curse of dimensionality problem in the mathematics and engineering literature. In contrast, little is known on how the human brain solves problems of this kind with relative ease. The key question is: How does the brain do it? To address this basic problem, it may be that we can learn from the mathematics and engineering literature, reformulated in the context of neuroscience. This one-day workshop at NIPS 2009 is aimed at addressing the issues involved in the curses (and blessings) of dimensionality. Invited Speakers: Les Valiant (Harvard) Dr. John Tsotsos (York) Bill Bialek (Princeton) Naftali Tishby (Weizmann) Warren Powell (Princeton) Workshop Format: Each speaker will have 90 minutes including discussion. The talks are informal with interruptions welcome during the talks. ----- From holroyd at uvic.ca Sat Oct 10 13:35:30 2009 From: holroyd at uvic.ca (Clay Holroyd) Date: Sat, 10 Oct 2009 10:35:30 -0700 Subject: Connectionists: Tenure-track Position in Cognitive Neuroscience Message-ID: Tenure-track Position in Cognitive Neuroscience The Department of Psychology at the University of Victoria invites applications for a tenure-track appointment in cognitive neuroscience at the level of Assistant Professor to begin July 1st, 2010. The ideal candidate will have a strong background and training in functional magnetic resonance imaging and in computational cognitive neuroscience techniques, but applications demonstrating competence in either of these methods will be considered. The candidate?s research interests will add to the existing strengths of the department that include the study of cognitive control, attention, learning, memory, eyewitness testimony, visual cognition, motor control, development and aging. The successful candidate will play a key role in the development of a 1.5 T General Electric MRI system, presently being upgraded to be capable of acquiring functional data, at the Royal Jubilee Hospital located approximately 1 mile from campus. The candidate will also have access to Westgrid, a high-performance computing network that encompasses 14 partner institutions across four provinces. Duties include the maintenance of a successful program of research (as evidenced by publications and external grant support), teaching at the undergraduate and graduate levels and contributions to the collegiality, reputation and day-to-day operation of the Department and the University (e.g., collaborative research, curriculum development and committee service). Applicants must have a PhD in cognitive psychology, cognitive neuroscience, neuroscience, computer science, or a related discipline. To apply, submit a letter of application (including a statement of research interests and accomplishments and a statement of teaching interests and experience), a curriculum vitae, copies of publications, evidence of teaching experience, and arrange for three confidential letters of reference to be sent to: Cognitive and Brain Sciences Search Committee Department of Psychology University of Victoria PO Box 3050 STN CSC Victoria, British Columbia V8W 3P5 Canada The deadline is December 7, 2009, but applications will be reviewed until the position is filled. The University of Victoria is an equity employer and encourages applications from women, persons with disabilities, visible minorities, aboriginal peoples, persons of all sexual orientations and genders, and others who may contribute to the further diversification of the university. All qualified candidates are encouraged to apply; however, in accordance with Canadian immigration requirements, Canadian and permanent residents will be given priority. The University of Victoria is located on beautiful Vancouver Island, in close proximity to Vancouver, BC, and Seattle, WA. Victoria is the provincial capital of British Columbia and provides a wealth of cultural and outdoor recreation opportunities. -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091010/ed56f299/attachment-0001.html From rmcantin at isr.ist.utl.pt Tue Oct 13 10:48:11 2009 From: rmcantin at isr.ist.utl.pt (Ruben Martinez-Cantin) Date: Tue, 13 Oct 2009 15:48:11 +0100 Subject: Connectionists: CfP: NIPS 09 Workshop on Adaptive Sensing, Active Learning and Experimental Design In-Reply-To: <24da2db50910130736l17041b7fn29737d195693cc8c@mail.gmail.com> References: <24da2db50909210758j37843345wa44c6c655bcd39ae@mail.gmail.com> <24da2db50910120704v407f79dyf82c4ffb99bfa5a3@mail.gmail.com> <24da2db50910130736l17041b7fn29737d195693cc8c@mail.gmail.com> Message-ID: <24da2db50910130748i559cd6fbo76dcac8dfe776c5f@mail.gmail.com> ---------------------------------------------------------------- ? ? ? ? ? ? ? ?CALL FOR CONTRIBUTIONS ? ? ? ? ? ? ? ? ? NIPS workshop on ?Adaptive Sensing, Active Learning and Experimental Design: ? ? ? ? ? Theory, Methods and Applications ? ? ? Whistler, BC, Canada, December 11, 2009 ? ? http://users.isr.ist.utl.pt/~rmcantin/nips2009.php ---------------------------------------------------------------- Important Dates: ---------------- ? * Submission of extended abstracts: October 27, 2009 ? ? ? (later submission might not be considered for review) ? * Notification of acceptance: November 5, 2009 ? * Workshop date: December 11, 2009 Overview: --------- The fields of active learning, adaptive sensing and sequential experimental design have seen a growing interest over the last decades in a number of communities, ranging from machine learning and statistics to biology and computer vision. Broadly speaking, all active and adaptive approaches focus on closing the loop between data analysis and acquisition. Said in a different way the goal is to use information collected in past samples to adjust and improve the future sampling and learning processes, in the spirit of the twenty questions game. These fields typically address the problem in very diverse ways, and using different problem formulations. The main objective of this workshop is to bring these communities together, share ideas and knowledge, and cross-fertilize the various fields. Invited speakers (confirmed): ----------------------------- ? * Maria-Florina Balcan, Georgia-Tech ? * Donald R. Jones. General Motors Corporation. ? * Andreas Krausse. Caltech ? * Dan Lizotte, University of Michigan ? * Luis Montesano, University of Zaragoza ? * Liam Paninski, Columbia University ? * Matthias Seeger, Saarland University, Saarbruecken Submission instructions: ------------------------ We invite submission of extended abstracts to the workshop. Extended abstracts should beat most 3 pages in length, formatted in according to NIPS style. However, the submission should not be blind. Extended abstracts should be sent in PDF or PS file format by email to alnips09 at gmail.com The selected submission may be accepted either as an oral presentation or as a poster presentation. We encourage participants who can contribute in the following areas: ? * Active learning ? * Active filtering ? * Sequential experimental design ? * Adaptive sensing ? * Optimal information gathering ? * Bayesian optimization ? * Active cognitive development ? * Robotics exploration ? * Sensor placement ? * Active signal processing ? * Online decision making ? * Active model discrimination. ? * Selection criteria/Utility functions ? * Information theoretic metrics The above list is not exhaustive, and we welcome submissions on highly related topics too. Accepted extended abstracts will be made available online at the workshop website. Organizers: ----------- ? * Rui Castro, Columbia University. ? * Nando de Freitas, University of British Columbia. ? * Ruben Martinez-Cantin , Instituto Superior Tecnico. Contact: -------- mailto:alnips09 at gmail.com http://users.isr.ist.utl.pt/~rmcantin/nips2009.php From nicolas.brunel at univ-paris5.fr Tue Oct 13 10:56:24 2009 From: nicolas.brunel at univ-paris5.fr (Nicolas Brunel) Date: Tue, 13 Oct 2009 16:56:24 +0200 Subject: Connectionists: Post-doctoral position in computational neuroscience Message-ID: <4AD49518.6030800@univ-paris5.fr> A postdoctoral position in computational neuroscience is available in Institute for Scientific Interchange (ISI) in Torino (Italy), to work on a research project in collaboration with Nicolas Brunel (Paris), Stefano Panzeri (Genova) and Nikos Logothetis (Tubingen). The goal of the project is to better understand the role of network oscillations in transmitting sensory information, through a combination of data analysis using information theoretic techniques, and analysis of the dynamics of models of networks of spiking neurons. This position demands an energetic and highly-motivated candidate with a strong background in computational/theoretical neuroscience and with the ability to work in an autonomous fashion. The successful candidate will be based in ISI, Torino, in the Statistical Physics research division (see www.isi.it). Frequent trips to Paris (www.neurophys.biomedicale.univ-paris5.fr/~brunel) Genova (www.iit.it/en/robotics-brain-and-cognitive-sciences/people.html?view=profile&layout=profile&id=289), and Tubingen (www.kyb.mpg.de/~nikos) are expected during the project. The position can start immediately, and is funded until July 2011. Salary will be commensurate with level of experience. Applications will be accepted until the position is filled. To apply, please send a CV and contact details of two or three references to nicolas.brunel at parisdescartes.fr -- Laboratory of Neurophysics and Physiology Universite Paris Descartes, CNRS UMR 8119 45 rue des Saints Peres 75270 Paris Cedex 06 Tel (33).1.42.86.20.58 - Fax (33).1.49.27.90.62 nicolas.brunel at univ-paris5.fr www.neurophys.biomedicale.univ-paris5.fr/~brunel From shivani at MIT.EDU Tue Oct 13 13:21:44 2009 From: shivani at MIT.EDU (Shivani Agarwal) Date: Tue, 13 Oct 2009 13:21:44 -0400 (EDT) Subject: Connectionists: 2nd CFP: NIPS 2009 Workshop - Advances in Ranking Message-ID: ************************************************************************ 2nd CALL FOR PAPERS ---- Advances in Ranking ---- Workshop at the 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009) http://web.mit.edu/shivani/www/Ranking-NIPS-09/ -- Submission Deadline: October 23, 2009 -- ************************************************************************ [ Apologies for multiple postings ] OVERVIEW -------- Ranking problems are increasingly recognized as a new class of statistical learning problems that are distinct from the classical learning problems of classification and regression. Such problems arise in a wide variety of domains: in information retrieval, one wants to rank documents according to relevance to a query; in natural language processing, one wants to rank alternative parses or translations of a sentence; in collaborative filtering, one wants to rank items according to a user's likes and dislikes; in computational biology, one wants to rank genes according to relevance to a disease. Consequently, there has been much interest in ranking in recent years, with a variety of methods being developed and a whole host of new applications being discovered. This workshop aims to bring together researchers interested in the area to share their perspectives, identify persisting challenges as well as opportunities for meaningful dialogue and collaboration, and to discuss possible directions for further advances and applications in the future. One of the primary goals of the workshop will be to reach out to a broad audience. To this end, we will have talks on topics ranging from more statistically/mathematically oriented approaches to ranking, to newer application areas. A second goal will be to bring to the fore a range of questions that are currently being debated within the community, for example via a panel discussion between experts in the field. Overall, the workshop will aim to provide a forum that showcases recent advances in ranking to the broader community, facilitates open debate on some of the questions in this area, and helps catalyze further interest among those new to the topic. FORMAT ------ This is a one-day workshop that will follow the 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009). The workshop will consist of two 3-hour sessions. There will be 2-4 invited talks by experts in the area, and 5-8 short talks. Depending on interest and submissions, we may also have a short poster/demo session. There will be time at the end of each talk/session for questions and discussion. We also intend to have a panel discussion that will be moderated by the organizers and that will bring together leading figures from both academia and industry for discussion and debate. Keynote Lecture --------------- * Persi Diaconis, Stanford University Invited Talks ------------- * Ralf Herbrich, Microsoft Research Cambridge * Lek-Heng Lim, University of California, Berkeley Contributed Talks ----------------- These will be based on submissions of short papers. See below for details. CALL FOR PAPERS --------------- We invite submissions of short papers addressing all aspects of ranking in machine learning, data mining, and statistics, as well as related application areas. These include for example: * algorithms for ranking * mathematical/statistical analyses of ranking * relationships between ranking and other problems * applications of ranking in information retrieval, natural language processing, collaborative filtering, computational biology, computer vision, and other areas * benchmark data sets for ranking * challenges in applying or analyzing ranking methods We also welcome papers on ranking that do not fit into one of the above categories, as well as papers that describe work in progress. Please note that papers that have previously appeared (or have been accepted for publication) in a journal or at a conference or workshop, or that are being submitted to another workshop, are not appropriate for this workshop. All papers presented at the workshop will be made available as electronic proceedings. A print version will be distributed at the workshop. Submission Instructions ----------------------- Submissions should be upto 4 pages in length using NIPS style files (available at http://web.mit.edu/shivani/www/Ranking-NIPS-09/StyleFiles/), and should include the title, authors' names, postal and email addresses, and a brief abstract. Email submissions (in pdf or ps format only) to shivani at mit.edu with subject line "Workshop Paper Submission". The deadline for submissions is Friday October 23, 5:00 pm EDT. Submissions will be reviewed by the program committee and authors will be notified of acceptance/rejection decisions by Wednesday November 11. Final versions of accepted papers will be due on Wednesday November 18. Please note that one author of each accepted paper must be available to present the paper at the workshop. RELATED WORKSHOP ---------------- There is a related workshop at NIPS this year titled Learning with Orderings, with a somewhat different focus than ours; see the webpage above for details. We encourage all participants of our workshop to attend this workshop as well. Submissions should be directed to only one workshop; if a submission cannot be accommodated by one workshop, it may be forwarded to the other workshop for consideration. Please indicate if you would like us to consider this. IMPORTANT DATES --------------- First call for papers -- September 15, 2009 Paper submission deadline -- October 23, 2009 (5:00 pm EDT) Notification of decisions -- November 11, 2009 Final papers due -- November 18, 2009 Workshop -- December 11, 2009 ORGANIZERS ---------- * Shivani Agarwal, MIT * Chris J.C. Burges, Microsoft Research * Koby Crammer, The Technion CONTACT ------- Please direct any questions to shivani at mit.edu. ************************************************************************ From murphyk at cs.ubc.ca Tue Oct 13 15:11:18 2009 From: murphyk at cs.ubc.ca (Kevin Murphy) Date: Tue, 13 Oct 2009 12:11:18 -0700 (PDT) Subject: Connectionists: postdoc at UBC in vision/learning Message-ID: I am pleased to announce that the Canadian Institute For Advanced Research (CIFAR) is funding a 2 year Junior Fellow position as part of its Neural Computation and Adaptive Perception (NCAP) program. The Fellow will be based at the University of British Columbia (UBC) in Vancouver, and will work with one or more of Kevin Murphy, Nando de Freitas, and David Lowe. The starting date is flexible, but it is expected to be sometime between June and September 2010. The precise topics of research are still to be determined, but will be related to using unsupervised, semi-supervised and 'deep learning' to solve interesting vision problems, such as generic object detection and localisation, image segmentation and labeling, video parsing, feature discovery, active vision, etc. The ideal candidate will have a strong research track record in learning and vision, and an interest in neuroscience, as well as excellent communication skills, and a strong potential to collaborate with other NCAP program members. More details on the NCAP program can be found here: http://www2.cifar.ca/research/neural-computation-and-adaptive-perception-program/ The candidate will be expected to fully participate in NCAP meetings (usually 2-3 per year, typically in Vancouver or Toronto), and CIFARs Junior Fellow Academy, interacting with peers in CIFAR's other research programs. More details on the Junior Academy can be found here: http://www2.cifar.ca/research/junior-academy/ Please send a 2 page research statement and a recent CV to murphyk at cs.ubc.ca by 15 Jan 2010. Also, please arrange to have 3 letters of reference sent to the same address, by 15 Jan 2010. These letters cannot be from CIFAR members. (You can have additional letters of reference from CIFAR members sent if you wish.) Kevin Murphy From juffi at ke.informatik.tu-darmstadt.de Mon Oct 12 16:58:22 2009 From: juffi at ke.informatik.tu-darmstadt.de (Johannes Fuernkranz) Date: Mon, 12 Oct 2009 22:58:22 +0200 Subject: Connectionists: CfP ICML 2010 Haifa Message-ID: [Please distribute, apologies for multiple postings] ----------------------------------------------------------------------- ICML 2010 Call for Papers ----------------------------------------------------------------------- The 27th International Conference on Machine Learning (ICML-10) will be held in Haifa, Israel, June 21-25, 2010, and will be co-located with the 23rd Conference on Learning Theory (COLT-10). ICML 2010 invites submission of engagingly written papers on substantial, original, and previously unpublished research in all aspects of machine learning. We welcome submissions of innovative work on systems that are self adaptive, systems that improve their own performance, or systems that apply logical, statistical, probabilistic or other formalisms to the analysis of data, to the learning of predictive models, to cognition, or to interaction with the environment. We welcome innovative applications, theoretical contributions, carefully evaluated empirical studies, and we particularly welcome work that combines all of these elements. We also encourage submissions that bridge the gap between machine learning and other fields of research. To further strengthen and broaden participation in all areas of machine learning, we followed an area-driven process for selecting the area chairs and are confident that we have competent people for handling submissions in /all/ areas of machine learning. Further information is available at http://icml2010.haifa.il.ibm.com/. IMPORTANT DATES * October 12, 2009 Call for papers issued * February 1, 2010 Full paper submissions due (no separate abstract date) Deadline: 11:59pm, Samoa time * April 16, 2009 Acceptance notification * June 21, 2010 ICML tutorials * June 22-24, 2010 ICML conference * June 25, 2010 Joint ICML/COLT workshop day * June 26, 2010 Organized trip (Jerusalem or Nazareth) * June 27-29, 2010 COLT conference FORMAT OF THE CONFERENCE The conference will include three days of technical presentations, one day of tutorials and one day of workshops. Accepted papers will each have an oral presentation as well as a poster in an evening poster session. Awards will be given for papers of outstanding quality. There will also be talks by several invited speakers, and a banquet. SUBMISSION Submission of papers and the management of the paper reviewing process will be entirely electronic. More instructions for authors can be found at http://icml2010.haifa.il.ibm.com/. REVIEW PROCESS The review process will be similar to last year's process, but incorporates some of the community feedback. Authors, reviewers, and area chairs indicate subject areas. With the help of these subject areas, area chairs bid for papers and one area chair is assigned to each paper. During a first round of reviewing, each paper will receive two reviews. First-round reviewers are assigned via subject areas and bidding. As in recent years, authors will have the opportunity to see and respond to the reviews before a final decision is made. Papers that receive at least one positive review in the first round, or where otherwise deemed necessary by the area chair, will receive one or more additional reviews. These additional reviewers are manually selected by the area chair. Final decisions will be made using the input from all reviewers, the author feedback, the area chair, and the program chairs. Reviewing for ICML 2010 will be blind to the identities of the authors. No conditional accepts will be granted this year. ICML 2010 will not accept any paper that is substantially similar to another paper that is currently under review or has already been accepted for publication in a journal or another conference. Please clearly indicate in the submission which contributions are novel and which are previous work, either by the authors or others. If a paper submitted to ICML 2010 and another already published or already submitted paper contain substantial overlap in content and this overlap is not clearly indicated (anonymously) as being previous work, then the ICML submission may be rejected on the grounds of being a dual submission. Similarly, authors must withdraw their papers if they submit an overlapping paper elsewhere during ICML's review period. For papers published in substantially disjoint communities (application conferences, for example), the amount of novel content a paper needs to contain may be less, as long as the submitted papers are themselves clearly targeted to a machine-learning audience. With your help, we expect another excellent conference! THE ICML2010 ORGANIZATIONAL TEAM * General Chair: o Stefan Wrobel (Fraunhofer IAIS & Universität Bonn) * Program Co-chairs: o Johannes Fürnkranz (Technische Universität Darmstadt) o Thorsten Joachims (Cornell University) * Local Arrangements Chair: o Shai Fine (IBM Research Haifa) Contact us at From cardoso at bcos.uni-freiburg.de Tue Oct 13 05:24:05 2009 From: cardoso at bcos.uni-freiburg.de (Simone Cardoso de Oliveira) Date: Tue, 13 Oct 2009 11:24:05 +0200 Subject: Connectionists: Over 30 positions in Computational Neuroscience in Germany - visit the Bernstein Network booth #2146 at SfN 2009! Message-ID: <4AD44735.7080309@bcos.uni-freiburg.de> Dear colleagues, at the upcoming SfN meeting in Chicago, the German Bernstein Network Computational Neuroscience presents itself with - more than 30 job offers, - more than 20 study programs, - at more than 20 locations in Germany! In addition, the booth features demos of: - EyeSeeCam -- a novel head mounted camera controlled by the user's eye movements - see through someone else's eyes! (Sunday, October 18 - Wednesday, October 21 at 9:40am, 1:00pm, and 4:00pm) - a new MRI atlas of the rhesus monkey brain (Roland Tammer, Sabine Hofer, Klaus-Dietmar Merboldt, Jens Frahm: "Magnetic Resonance Imaging of the Rhesus Monkey Brain", Vandenhoeck&Ruprecht) We look forward to welcoming you at booth #2146! Best regards, Kerstin Schwarzwaelder (Bernstein Coordination Site representative at the booth) Simone Cardoso (Head of the Bernstein Coordination Site) -- Dr. Simone Cardoso de Oliveira Bernstein Network Computational Neuroscience Head of the Bernstein Coordination Site (BCOS) Albert Ludwigs University Freiburg Hansastr. 9A 79104 Freiburg, Germany phone: +49-761-203-9583 fax: +49-761-203-9585 cardoso at bcos.uni-freiburg.de www.nncn.de From mpullano at MIT.EDU Tue Oct 13 09:37:55 2009 From: mpullano at MIT.EDU (Michelle M Pullano) Date: Tue, 13 Oct 2009 09:37:55 -0400 Subject: Connectionists: Koller and Friedman/Probabilistic Graphical Models from the MIT Press Message-ID: The MIT Press is pleased to announce the publication of Probabilistic Graphical Models: Principles and Techniques, by Daphne Koller and Nir Friedman. http://mitpress.mit.edu/9780262013192 A general framework for constructing and using probabilistic models of complex systems that would enable a computer to use available information for making decisions 2009 ? 1208 pp., 399 illus. ? hardcover ? $95.00/?62.95 ? 978-0-262-01319-2 "This landmark book provides a very extensive coverage of the field, ranging from basic representational issues to the latest techniques for approximate inference and learning. As such, it is likely to become a definitive reference for all those who work in this area. Detailed worked examples and case studies also make the book accessible to students." --Kevin Murphy, Department of Computer Science, University of British Columbia -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091013/1d99a45b/attachment-0001.html From mail at jan-peters.net Thu Oct 15 09:35:40 2009 From: mail at jan-peters.net (Jan Peters) Date: Thu, 15 Oct 2009 15:35:40 +0200 Subject: Connectionists: PostDoc Position: Learning to Grasp with Motor Primitives Message-ID: We are searching for a PostDoc at the Robot Learning Lab of the Department of Empirical Inference, at the Max Planck Institute for Biological Cybernetics, T?bingen, Germany. If you or one of your students is graduating within the next five-six month and is searching for a position that involves robot grasping, application of machine learning in robotics and motor primitive learning, then please contact Jan Peters (mail at jan-peters.net) informally by email! For more information on the institute see http://www.kyb.tuebingen.mpg.de and on our past projects in robot learning see http://www.robot-learning.de or http://www.jan-peters.net Best wishes, Jan Peters PostDoc Position ============= The Department of Empirical Inference, headed by Bernhard Sch?lkopf, at the Max Planck Institute for Biological Cybernetics, T?bingen, Germany, intends to fill an open Post-Doc position in the EU Project GeRT. Applications are encouraged in area of robot learning. We are looking for persons with a strong background in one or several of the following areas: * robot grasping * application of machine learning in robotics * motor primitive learning The Post-Doc will work on the EU Project GeRT in the robot learning group headed by Jan Peters for a duration of up to 3 years. Latest date for application submission is January 15th. Applicants should hold a PhD in computer science, engineering or related fields and have an outstanding publication record at the relevant conferences (e.g., ICRA, IROS, R:SS, NIPS) and journals (e.g., IJRR, IEEE TRO, Autonomous Robots, JMLR, Machine Learning). We are looking for highly motivated and creative individuals who enjoy collaborating in a team of researchers with various cultural and scientific backgrounds. Applicants should be capable of conducting independent research, willing to (co-)supervise PhD students, and contribute to the success of our group as a whole. The working language in our group is English. The Department for Empirical Inference is a young and energetic machine learning group. The Max Planck Institute for Biological Cybernetics provides an excellent research environment in all respects. The robot learning lab is present both in the core robotics and in the machine learning community. Applications should include a statement of research experience and interests, a CV, and contact details of at least two referees. Please send your applications in electronic form to sabrina.nielebock at tuebingen.mpg.de . The Max Planck Society is an equal opportunity employer and is committed to employing more handicapped individuals and especially encourages them to apply. The Max Planck Society wishes to increase the share of women in areas in which they are underrepresented. Women are strongly invited to apply. From mohri at cs.nyu.edu Thu Oct 15 13:51:18 2009 From: mohri at cs.nyu.edu (Mehryar Mohri) Date: Thu, 15 Oct 2009 13:51:18 -0400 Subject: Connectionists: COLT 2010 - Preliminary Call for Papers Message-ID: <2ED0AE6E-9E5C-47EF-AF27-31D07D384578@cs.nyu.edu> COLT 2010 - Call for Papers The 23rd Annual Conference on Learning Theory (COLT 2010) will take place in Haifa, Israel, on June 27-29, 2010 and will be co-located with ICML 2010. We invite submissions of papers addressing theoretical aspects of machine learning and empirical inference. We strongly support a broad definition of learning theory, including: Analysis of learning algorithms and their generalization ability Computational complexity of learning Bayesian analysis Statistical mechanics of learning systems Optimization procedures for learning Kernel methods Inductive inference Boolean function learning Unsupervised and semi-supervised learning and clustering On-line learning and relative loss bounds Learning in planning and control, including reinforcement learning Learning in games, multi-agent learning Mathematical analysis of learning in related fields, e.g., game theory, natural language processing, neuroscience, bioinformatics, privacy and security, machine vision, data mining, information retrieval We are also interested in papers that include viewpoints that are new to the COLT community. We welcome experimental and algorithmic papers provided they are relevant to the focus of the conference by elucidating theoretical results in learning. Also, while the primary focus of the conference is theoretical, papers can be strengthened by the inclusion of relevant experimental results. Papers that have previously appeared in journals or at other conferences, or that are being submitted to other conferences, are not appropriate for COLT. Papers that include work that has already been submitted for journal publication may be submitted to COLT, as long as the papers have not been accepted for publication by the COLT submission deadline (conditionally or otherwise) and that the paper is not expected to be published before the COLT conference (June 2010). Feedback on Review Quality There will be no rebuttal phase this year. However, authors will be given the opportunity to assess the quality of reviews and provide feedback to the reviewers, after the decisions have been made. These assessments will be used in particular to determine the Best Reviewer award (see below). Paper and Reviewer Awards This year, COLT will award both best paper and best student paper awards. Best student papers must be authored or coauthored by a student. Authors must indicate at submission time if they wish their paper to be eligible for a student award. This does not preclude the paper to be eligible for the best paper award. To further emphasize the importance of the reviewing quality, this year, COLT will also award a best reviewer award to the reviewer who has provided the most insightful and useful comments. Open Problems Session We also invite submission of open problems (see separate call). These should be constrained to two pages. There is a shorter reviewing period for the open problems. Accepted contributions will be allocated short presentation slots in a special open problems session and will be allowed two pages each in the proceedings. Paper Format and Electronic Submission Instructions Formatting and submission instructions will be available in early December at the conference website. Important Dates Preliminary call for papers issued October 15, 2009 Electronic submission of papers (due by 5:59pm PST) February 19, 2010 Electronic submission of open problems March 13, 2010 Notice of acceptance or rejection May 07, 2010 Submission of final version May 21, 2010 Feedback on reviews due May 28, 2010 Joint ICML/COLT workshop day June 25, 2010 2010 COLT conference June 27-29, 2010 Organization Program Co-chairs: Adam Tauman Kalai (Microsoft Research) Mehryar Mohri (Courant Institute of Mathematical Sciences and Google Research) Program Committee: Shivani Agarwal Mikhail Belkin Shai Ben-David Nicol? Cesa-Bianchi Ofer Dekel Steve Hanneke Jeff Jackson Sham Kakade Vladimir Koltchinskii Katrina Ligett Phil Long Gabor Lugosi Ulrike von Luxburg Yishay Mansour Ryan O?Donnell Massimiliano Pontil Robert Schapire Rocco Servedio John Shawe-Taylor Shai Shalev-Shwartz Gilles Stoltz Ambuj Tewari Jenn Wortman Vaughan Santosh Vempala Manfred Warmuth Robert Williamson Thomas Zeugmann Tong Zhang Local Arrangements Chair: Shai Fine (IBM Research Haifa) Invited speakers Prof. Noga Alon - School of Mathematical Sciences, Tel Aviv University Prof. Noam Nisan - School of Computer Science and Engineering, The Hebrew University Jerusalem -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091015/b01af10d/attachment-0001.html From pascal.fua at epfl.ch Fri Oct 16 04:12:39 2009 From: pascal.fua at epfl.ch (Pascal Fua) Date: Fri, 16 Oct 2009 10:12:39 +0200 Subject: Connectionists: Open Faculty Positions at EPFL Message-ID: <4AD82AF7.4040101@epfl.ch> The School of Computer and Communication Sciences at EPFL invites applications for faculty positions in computer science. We are primarily seeking candidates for tenure-track assistant professor positions, but suitably qualified candidates for senior positions will also be considered. Successful candidates will develop an independent and creative research program, participate in both undergraduate and graduate teaching, and supervise PhD students. Candidates from all areas of computer science will be considered, but preference will be given to candidates with interests in algorithms, bio-informatics, machine learning and verification. EPFL offers internationally competitive salaries, significant start-up resources, and outstanding research infrastructure. To apply, please follow the application procedure at http://icrecruiting.epfl.ch. The following documents are requested in PDF format: curriculum vitae, including publication list, brief statements of research and teaching interests, names and addresses (including e-mail) of 3 references for junior positions, and 6 for senior positions. Screening will start on January 15, 2010. Further questions can be addressed to: Professor Willy Zwaenepoel Dean School of Computer and Communication Sciences EPFL CH-1015 Lausanne, Switzerland recruiting.ic at epfl.ch For additional information on EPFL, please consult: http://www.epfl.ch or http://ic.epfl.ch EPFL is an equal opportunity employer. -- -------------------------------------------------------------------- Prof. P. Fua (Pascal.Fua at epfl.ch) Tel: 41/21-693-7519 FAX: 41/21-693-7520 Url: http://cvlab.epfl.ch/~fua/ -------------------------------------------------------------------- From jean-pascal.pfister at eng.cam.ac.uk Fri Oct 16 10:03:25 2009 From: jean-pascal.pfister at eng.cam.ac.uk (Jean-Pascal Pfister) Date: Fri, 16 Oct 2009 15:03:25 +0100 Subject: Connectionists: Call for abstract: NIPS 2009 workshop on normative electrophysiology. Message-ID: Please note that the submission deadline has been extended to October 30th. ------------------------------- CALL FOR ABSTRACT ------------------------------- *NIPS 2009 WORKSHOP ON NORMATIVE ELECTROPHYSIOLOGY* We are now soliciting abstracts (see format below) for the NIPS 2009 Workshop on : Normative Electrophysiology: explaining cellular properties of neurons from first principles. Authors of accepted abstracts will be entitled to present a poster during the workshop. *Webpage* http://learning.eng.cam.ac.uk/Public/Lengyel/EventNips09 *Key dates* - abstract submission deadline: October 30th, 2009 - Notification of acceptance: November 5th, 2009 - Workshop: December 11th, 2009 *Workshop description* In the past decades, computational neuroscience has seen a burgeoning of normative approaches. These studies made significant advances in formulating formal theories of optimality, and optimal computations, identifying relevant physical and computational constraints under which those computations need to be implemented, developing analytical methods and numerical algorithms to solve the resulting constrained optimization problems, and relating these solutions to biological substrates. However, only a relatively small fraction of these studies attempted to make specific predictions about, and thus interpret in normative terms, the cellular-level electrophysiological properties of individual neurons or synapses. Small in numbers it may be, the potential impact of this particular line of research cannot be ignored as such theories may provide a way to bridge the gap between the cellular-molecular and the systems-level branches of neuroscience by connecting low-level properties of the nervous system to its high-level functions. Our workshop aims to highlight and discuss recent work in this field. Since much of the theoretical background in this field has been adopted from information theory, machine learning, and related fields, we expect that not only experimental and computational neuroscientists, but also machine learning researchers will be interested in the general topic and the specific talks. *Speakers* - Sophie Den?ve, ?cole Normale Sup?rieure - Adrienne Fairhall, U Washington - Aldo Faisal, Imperial College London - M?t? Lengyel, U Cambridge - Jean-Pascal Pfister, U Cambridge - Tatyana Sharpee, Salk Institute - Taro Toyoizumi, Columbia U *Workshop location* Westin Resort and Spa / Hilton Whistler Resort and Spa Whistler, B.C., Canada *Submission instructions* Please submit abstracts (maximum 300 words) in plain text format by email directly to jean-pascal.pfister at eng.cam.ac.uk with the mention "Normative Electrophysiology". *Organizers* Jean-Pascal Pfister (primary contact) Computational and Biological Learning Lab Department of Engineering University of Cambridge Trumpington Street, Cambridge CB2 1PZ United Kingdom tel: +44 (0)1223 748 506 fax: +44 (0)1223 332 662 e-mail: jean-pascal.pfister at eng.cam.ac.uk M?t? Lengyel Computational and Biological Learning Lab Department of Engineering University of Cambridge Trumpington Street, Cambridge CB2 1PZ United Kingdom tel: +44 (0)1223 748 532 fax: +44 (0)1223 332 662 e-mail: m.lengyel at eng.cam.ac.uk -- Jean-Pascal Pfister, PhD Computational and Biological Learning Lab Cambridge University Engineering Department Trumpington Street, Cambridge CB2 1PZ, UK tel: +44 (0)1223 748 506, fax: +44 (0)1223 332 662 email: jean-pascal.pfister at eng.cam.ac.uk http://www.eng.cam.ac.uk/~jptp2/ -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091016/5167a4f6/attachment.html From chiestand at salk.edu Fri Oct 16 21:00:33 2009 From: chiestand at salk.edu (Chris Hiestand) Date: Fri, 16 Oct 2009 18:00:33 -0700 Subject: Connectionists: Neural Information Processing Systems Conference Registration Is Open Message-ID: <78B0A1E2-0BFD-4448-AC95-9ACBBE8EC175@salk.edu> Registration for the 2009 Neural Information Processing Systems Conference is open http://nips.cc/Register/ The last day for early registration pricing is November 6, 2009. Tutorials: December 7, 2009 in Vancouver, British Columbia, Canada Main Conference: December 7 - 10, 2009 in Vancouver, British Columbia, Canada Workshops: December 11-12, 2009 in Whistler, British Columbia, Canada The tutorials will offer a choice of six two-hour tutorials by leading scientists. The topics span a wide range of subjects including Neuroscience, Learning Algorithms and Theory, Bioinformatics, Image Processing, and Data Mining. The NIPS Conference features a single track program, with contributions from a large number of intellectual communities. Presentation topics include: Algorithms and Architectures; Applications; Brain Imaging; Cognitive Science and Artificial Intelligence; Control and Reinforcement Learning; Emerging Technologies; Learning Theory; Neuroscience; Speech and Signal Processing; and Visual Processing. All papers are rigorously reviewed. The Poster Sessions will take place Monday through Wednesday evenings, December 7 ? 9, 2009, during the Conference. The sessions offer high- quality posters and an opportunity for researchers to share their work and exchange ideas in a collegial setting. The majority of contributions accepted at NIPS are presented as posters. The Demonstrations enable researchers to highlight scientific advances, systems, and technologies in ways that go beyond conventional poster presentations. It provides a unique forum for demonstrating advanced technologies ? both hardware and software ? and fostering the direct exchange of knowledge. Following the regular program of the Neural Information Processing Systems 2009 conference in Vancouver, BC, Canada, four mini-symposia will be held in parallel during the afternoon of Thursday, December 10, 2009, in the Hyatt Regency, Vancouver, BC, Canada. The Post-Conference Workshop Program takes place in Whistler, B.C. at the Westin Resort and Spa and the Hilton Whistler Resort and Spa Friday, December 11 and Saturday, December 12, 2009. There will be 28 workshops covering a wide range of topics from Neuroscience to Machine Learning. The workshop program schedule allows time for informal discussions, skiing and other winter sports. See the NIPS Conference Program here: http://nips.cc/Conferences/2009/Program/ http://nips.cc -------------- next part -------------- A non-text attachment was scrubbed... Name: smime.p7s Type: application/pkcs7-signature Size: 2419 bytes Desc: not available Url : https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091016/ff1c3d11/smime.bin From serre at mit.edu Sat Oct 17 17:22:44 2009 From: serre at mit.edu (Thomas R.G. Serre) Date: Sat, 17 Oct 2009 16:22:44 -0500 Subject: Connectionists: Postdoc and research assistant positions, Department of Cognitive & Linguistics Science, Brown University (Providence, RI) Message-ID: Dear colleagues, I will be joining the Department of Cognitive & Linguistics Science at Brown University (Providence, RI) as an Assistant Professor in January 2010 and will be recruiting a full-time research assistant and a postdoc to start immediately. I would be grateful if you could please forward the advertisement below along to anyone you think would be appropriate. Best regards, --Thomas Serre Post-doctorate Associate - MIT - Brain & Cognitive Sciences Tel: +1 617 253 0548 | GMT -5:00 | http://web.mit.edu/serre/www/ ***************************************** Postdoctoral Research Associate, Department of Cognitive & Linguistics Science, Brown University (Providence, RI). We seek a talented postdoc to work as part of an interdisciplinary team to study the brain mechanisms underlying the recognition of objects and complex visual scenes. The goal of the project is to study how categorical decisions are learned and how objects are represented in the human brain using a combination of behavioral, imaging and physiological techniques, which are supported by rigorous computational models. In particular, we will use detailed quantitative computational models of object recognition to derive quantitative receptive-fields to characterize the relationship between visual stimuli and fMRI activity in higher visual areas. Reliable ?mind reading? from fMRI data will be used to validate the underlying computational models. The successful candidate must hold a Ph.D. in Psychology, Neuroscience, Cognitive Science, Computer Science, Engineering, or other related field and a strong research track record in cognitive neuroscience; prior experience with machine learning and/or computational neuroscience techniques as well as fluency in at least one programming language (e.g., Matlab, Python, C/C++) would be considered a plus. Expected start date: January 2010. Review of applications will continue until the position is filled. Interested applicants should apply electronically to Thomas Serre at tserre at gmail.com with a subject line containing ?Application for postdoc position | fMRI?. Applicants should include a list of publications, a Curriculum Vitae and the names of three references in their electronic application (no letters required at this stage). I will be attending the Society for Neuroscience meeting in Chicago. Potential candidates are encouraged to contact me tserre at gmail.com to arrange a meeting. Our lab webpage is still under construction but general information about my research interests can be found at http://web.mit.edu/serre/www. The lab will be part of the newly formed Department of Cognitive, Linguistic & Psychological Sciences (?CLiPS?) dedicated to the multidisciplinary study of the mind, brain, behavior, and language, which will be housed (together with the Brown Institute for Brain Science) in a new 40,000+ sq ft building. ************ Full-time Research Assistant, Department of Cognitive & Linguistics Science, Brown University (Providence, RI). We seek a talented research assistant to work as part of an interdisciplinary team to study the brain mechanisms underlying the recognition of objects and complex visual scenes. The lab typically uses a combination of behavioral, imaging and physiological techniques, which are supported by rigorous computational models. The successful candidate will assist with all aspects of our lab?s research and in particular: programming experiments and recruiting participants, collecting and analyzing experimental data (eye tracking, fMRI, iEEG and monkey electrophysiology) as well as providing miscellaneous research support. The successful candidate must hold a bachelor?s degree in Psychology, Cognitive Science, Neuroscience, Computer Science, Engineering, Math, or other related field. The candidate is also expected to be proficient in at least one programming language (e.g., Matlab, Python, C/C++). This position is ideal for candidates who are planning to attend graduate school and want additional research experience. Expected start date: January 2010. Review of applications will continue until the position is filled. Interested applicants should apply electronically to Thomas Serre at tserre at gmail.com with a subject line containing ?Application for research assistant position | fMRI?. Applicants should include a list of publications, a Curriculum Vitae and the names of two references in their electronic application (no letters required at this stage). I will be attending the Society for Neuroscience meeting in Chicago. Potential candidates are encouraged to contact me tserre at gmail.com to arrange a meeting. Our lab webpage is still under construction but general information about my research interests can be found at http://web.mit.edu/serre/www. The lab will be part of the newly formed Department of Cognitive, Linguistic & Psychological Sciences (?CLiPS?) dedicated to the multidisciplinary study of the mind, brain, behavior, and language, which will be housed (together with the Brown Institute for Brain Science) in a new 40,000+ sq ft building. -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091017/eb6ebaf5/attachment-0001.html From padams at notes.cc.sunysb.edu Sun Oct 18 11:32:19 2009 From: padams at notes.cc.sunysb.edu (padams@notes.cc.sunysb.edu) Date: Sun, 18 Oct 2009 11:32:19 -0400 Subject: Connectionists: 2 new papers on Hebbian learning Message-ID: Dear Colleagues - Recently, it has been shown that biological Hebbian plasticity (e.g. ltp) is not completely synapse-specific. We would like to draw your attention to our 2 new papers examining the effect of such minor inaccuracy in 2 simple neural net learning paradigms, PCA and ICA. For the linear (PCA) learning rule, inaccuracy produces graceful degradation, but for nonlinear (ICA) rules it can have a catastrophic effect. We propose that a critical learning problem in the brain is achieving the necessary (but likely synaptically impossible) extraordinary accuracy. Citations and abstracts follow. 1. Radulescu, A., Cox, K., and Adams, P. (2009). Hebbian errors in learning: an analysis using the Oja model. J. Theor. Biol. 258, 489-501 "Recent work on long term potentiation in brain slices shows that Hebb's rule is not completely synapse-specific, probably due to intersynapse diffusion of calcium or other factors. We previously suggested that such errors in Hebbian learning might be analogous to mutations in evolution. We examine this proposal quantitatively, extending the classical Oja unsupervised model of learning by a single linear neuron to include Hebbian inspecificity. We introduce an error matrix E, which expresses possible crosstalk between updating at different connections. When there is no inspecificity, this gives the classical result of convergence to the first principal component of the input distribution (PC1). We show the modified algorithm converges to the leading eigenvector of the matrix EC, where C is the input covariance matrix. In the most biologically plausible case when there are no intrinsically privileged connections, E has diagonal elements Q and off-diagonal elements (1-Q)/(n-1), where Q, the quality, is expected to decrease with the number of inputs n and with a synaptic parameter b that reflects synapse density, calcium diffusion, etc. We study the dependence of the learning accuracy on b, n and the amount of input activity or correlation (analytically and computationally). We find that accuracy decreases (learning becomes gradually less useful) with increases in b, particularly for intermediate (i.e., biologically realistic) correlation strength, although some useful learning always occurs up to the trivial limit Q=1/n. We discuss the relation of our results to Hebbian unsupervised learning in the brain. When the mechanism lacks specificity, the network fails to learn the expected, and typically most useful, result, especially when the input correlation is weak. Hebbian crosstalk would reflect the very high density of synapses along dendrites, and inevitably degrades learning." link: http://dx.doi.org/10.1016/j.jtbi.2009.01.036 2 Cox K.J.A. and Adams P.R. (2009) Hebbian crosstalk prevents nonlinear unsupervised learning. Front. Comput. Neurosci. 3:1-20 2009 "Learning is thought to occur by localized, activity-induced changes in the strength of synaptic connections between neurons. Recent work has shown that activity-dependent changes at one connection can affect changes at others ("crosstalk"). We studied the role of such crosstalk in nonlinear Hebbian learning using a neural network implementation of Independent Components Analysis (ICA). We find that there is a sudden qualitative change in the performance of the network at a threshold crosstalk level and discuss the implications of this for nonlinear learning from higher-order correlations in the neocortex." link: http://frontiersin.org/computationalneuroscience/paper/10.3389/neuro.10/011.2009/ Comments and feedback welcome - Paul Adams and Kingsley Cox -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091018/b30b6244/attachment.html From kirsch at bccn.uni-freiburg.de Mon Oct 19 08:14:01 2009 From: kirsch at bccn.uni-freiburg.de (Janina Kirsch) Date: Mon, 19 Oct 2009 14:14:01 +0200 Subject: Connectionists: PhD-Position "Investigation of activity-dependent signal integration in neocortical neurons" at the Bernstein Center Freiburg Message-ID: Phd-Position ?Investigation of activity-dependent signal integration in neocortical neurons? Our lab uses combined intra- and extracellular recordings in rat V1 in vivo, as well as dynamic photo stimulation of acute brain slices to study network dynamics in V1 and the influence of dynamical states on the integration of synaptic input in pyramidal cells. The offered position is funded by the EU FACETS program (http://facets.kip.uni-heidelberg.de), and should, therein, contribute to the aspect of experimental characterization of cortical cells and networks in vivo and in vitro. The goal of our work is to link the obtained electrophysiological data, in close collaboration with other groups at the BCCN and within FACETS, to new models of neocortical networks, aimed to better understand the mechanisms underlying network dynamics in the cortex. The phd position is available immediately for 3 years. We are looking for experimentalists with a solid background in electrophysiological recording techniques and interest in computational neuroscience. Please apply via the online application form: http://www.bccn2.uni-freiburg.de/p hd _applications/index.php (Project-ID: FACETS) -- Dr. Janina Kirsch -- Coordinator for the Teaching & Training Programs Bernstein Center Freiburg Albert-Ludwig University of Freiburg Hansastr. 9a D - 79104 Freiburg Germany Phone: +49 (0) 761 203-9575 Fax: +49 (0) 761 203-9559 Email: kirsch @bcf.uni-freiburg.de Web: www.bcf.uni-freiburg.de _____ -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091019/d5be92d1/attachment.html From dmodha at gmail.com Mon Oct 19 22:51:35 2009 From: dmodha at gmail.com (Dharmendra S Modha) Date: Mon, 19 Oct 2009 19:51:35 -0700 Subject: Connectionists: Career Opportunities in Cognitive Computing Message-ID: <001101ca5130$3ef00d80$bcd02880$@com> IBM has recently won Phase 1 of the DARPA SyNAPSE project that seeks to discover, demonstrate, and deliver algorithms of the brain via a combination of (computational) neuroscience, supercomputing, and nanotechnology. We are seeking world-class candidates with expertise in one or more of the following areas: computational neuroscience (spiking computation, synaptic plasticity, structural plasticity), reinforcement learning, nonlinear dynamical systems, systems of coupled difference equations, neuroanatomy (gray matter, white matter), neurophysiology, neuromodulation, network analysis, neuromorphic chip design, analog VLSI, digital VLSI, ultra low-power computing, asynchronous VLSI, address events, circuit simulation, chip layout, chip testing, large-scale simulations, MPI (message passing interface), programming distributed memory machines, visualization, and virtual environments (USARSim) for cognitive task design. Interdisciplinary candidates with background in computer science, electrical engineering, biomedical engineering, and computational neuroscience are strongly encouraged to apply. Outstanding communication skills, ability to interact with a large, technically diverse, distributed team, demonstrated publication record, and relentless focus on project metrics and milestones are a must. IBM is committed to creating a diverse environment and is proud to be an equal opportunity employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, gender, gender identity or expression, sexual orientation, national origin, genetics, disability, age, or veteran status. Research Scientist - https://jobs3.netmedia1.com/cp/job_summary.jsp?job_id=RES-0262561 Post Doc - https://jobs3.netmedia1.com/cp/job_summary.jsp?job_id=RES-0262571 Senior Engineer - https://jobs3.netmedia1.com/cp/job_summary.jsp?job_id=RES-0262576 -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091019/0d1554d4/attachment-0001.html From jaakko.peltonen at tkk.fi Tue Oct 20 13:11:19 2009 From: jaakko.peltonen at tkk.fi (jaakko.peltonen@tkk.fi) Date: Tue, 20 Oct 2009 20:11:19 +0300 (EEST) Subject: Connectionists: NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics, Final call for contributions Message-ID: ------------------------------------------------------------------ FINAL CALL FOR CONTRIBUTIONS NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics Whistler, BC, Canada, December 12, 2009 http://www.dcs.gla.ac.uk/~srogers/lms09/index.htm ------------------------------------------------------------------ Important Dates: ---------------- Submission of extended abstracts: October 27, 2009 Notification of acceptance: November 6, 2009 Workshop Description: --------------------- Learning from multiple sources denotes the problem of jointly learning from a set of (partially) related learning problems / views / tasks. This general concept underlies several subfields receiving increasing interest from the machine learning community, which differ in terms of the assumptions made about the dependency structure between learning problems. In particular, the concept includes topics such as data fusion, transfer learning, multitask learning, multiview learning, and learning under covariate shift. Several approaches for inferring and exploiting complex relationships between data sources have been presented, including both generative and discriminative approaches. The workshop will provide a unified forum for cutting edge research on learning from multiple sources; the workshop will examine the general concept, theory and methods, and will also examine *robotics* as a natural application domain for learning from multiple sources. The workshop will address methodological challenges in the different subtopics and further interaction between them. The intended audience is researchers working in fields of multi-modal learning, data fusion, and robotics. (More detailed background information is available at the workshop website.) The workshop includes a morning session focused on theory/methods, and an afternoon session focused on the robotics application. The workshop is a core event of the PASCAL2 Network of Excellence. PASCAL2 Invited Speakers: ------------------------- Morning Session: Chris Williams - University of Edinburgh Afternoon Session: Ingmar Posner - University of Oxford Submission Instructions: ------------------------ We invite submission of extended abstracts to the workshop. Extended abstracts should be 2-4 pages, formatted in the NIPS style: http://nips.cc/PaperInformation/StyleFiles Unlike the main NIPS conference, identities of authors do not need to be removed from the extended abstracts. Extended abstracts should be sent in .PDF or .PS file format by email, to either D.Hardoon at cs.ucl.ac.uk or gleen at cis.hut.fi. Acceptance to the workshop will be determined based on peer review of each extended abstract. Submissions are expected to represent high-quality, novel contributions in theory/methods of learning from multiple sources, or high-quality, novel contributions in application of learning from multiple sources to robotics (see below). To encourage participants from the machine learning community to test their algorithms in the domain of robotics, we provide a dataset, with computed features, representative of open research issues in robotics; see the workshop webpage for details. Robotics-oriented papers submitted to the workshop are strongly encouraged to contain an experimental evaluation on the database. The obtained results will be presented by the organizers during the workshop. Submitted extended abstracts may be accepted either as an oral presentation or as a poster presentation; there will be only a limited number of oral presentations in the morning and afternoon sessions. Accepted extended abstracts will be made available online at the workshop website. Depending on the quality of submissions, we will consider preparing a special issue of a journal or a collected volume on the topic of the workshop. A separate call for papers will then be issued after the workshop for the special issue/collected volume. Last year's "Learning from Multiple Sources" workshop led to a special issue in Machine Learning (currently in progress). Organisers ---------- * Barbara Caputo - Idiap Research Institute. * Nicolo Cesa-Bianchi - Universita?degli Studi di Milan. * David Hardoon - Institute for Infocomm Research (I2R). * Gayle Leen - Helsinki University of Technology. * Francesco Orabona - Idiap Research Institure. * Jaakko Peltonen - Helsinki University of Technology. * Simon Rogers - University of Glasgow. Programme Committee ------------------- * Cedric Archambeau - Xerox Research. * Andreas Argyriou - Toyota Technological Institute. * Claudio Gentile - Universita?dell'Insubria. * Mark Girolami - University of Glasgow. * Samuel Kaski - Helsinki University of Technology. * Arto Klami - Helsinki University of Technology. * John Shawe-Taylor - University College London. * Giorgio Valentini - Universita?degli Studi di Milan. Contact Persons --------------- For questions about the workshop, contact David R. Hardoon at D.Hardoon AT cs.ucl.ac.uk. From sumitb at microsoft.com Tue Oct 20 14:29:56 2009 From: sumitb at microsoft.com (Sumit Basu) Date: Tue, 20 Oct 2009 18:29:56 +0000 Subject: Connectionists: Final Call for Submissions: Workshop on Analysis and Design of Algorithms for Interactive Machine Learning at NIPS 2009 Message-ID: --------------------------------------------------------------------------------------------------------------------- ??? ADA-IML'09: Workshop on Analysis and Design of Algorithms for Interactive Machine Learning at NIPS 2009 ? ???????????????????????????????????????????? Saturday, 12/12/2009 ???????????????????????????? http://research.microsoft.com/~sumitb/adaiml09 ???????????????????? ***Deadline for abstract submission:? Friday, October 30, 5pm PST*** --------------------------------------------------------------------------------------------------------------------- INTRODUCTION AND GOALS OF THE WORKSHOP The traditional role of the human operator in machine learning problems is that of a batch labeler, whose work is done before the learning even begins.? However, there is an important class of problems in which the human is interacting directly with the learning algorithm as it learns.? Canonical problem scenarios which fall into this space include active learning, interactive clustering, query by selection, learning to rank, and others.? Such problems are characterized by three main factors: ??? 1. the algorithm requires input from the human during training, in the form of labels, feedback, parameter guidance, etc. ??? 2. the user cannot express an explicit loss function to optimize, either because it is impractical to label a large training ?????? set or because they can only express implicit preferences. ??? 3. the stopping criterion is performance that's "good enough" in the eyes of the user. The goal of this workshop is to focus on the machine learning techniques that apply to these problems, both in terms of surveying the major paradigms and sharing information about new work in this area. Through a combination of invited talks, discussions, and posters, we hope to gain a better understanding of the available algorithms and best practices for this space, as well as their inherent limitations. ? CALL FOR ABSTRACTS We invite all researchers interested in presenting at the workshop to submit a one-page abstract of their work.?? The presentation format will be a spotlight summary talk along with a poster session later in the afternoon.?? We encourage presentations on new, developing ideas, as well as previously published work the authors would like to discuss in this forum.? Feel free to email us if you are concerned about whether your work is appropriate for the workshop.?? Note that there will not be formal proceedings for the workshop, so authors need not be concerned about publishing work they present here at a later venue.? Please send your submissions (as PDF) by email to sumitb at microsoft dot com with the subject line "ADA-IML'09 Abstract: [submission title]" ***Deadline for submission:? Friday, October 30, 5pm PST*** INVITED SPEAKERS We have a number of invited speakers who will be presenting at the workshop, including:? -Jerry Zhu (University of Wisconsin, Madison) -Carlos Guestrin (CMU) -Rich Carauna (Microsoft Research) SCHEDULE Morning (7:30-10:30) ? I.??? Introduction (0.5 hr) II.?? Invited Talks (2 talks - 1.0 hr) III.? Poster preview talks (1.0 hr) IV.?Developing a Syllabus/Bibliography for Interactive Machine Learning (30 mins)* Afternoon (3:30-6:30) V.??? Invited Talks (2 talks - 1 hr) VI.?? Poster Session (1.5 hrs) VII.? Open Problems, Challenges, and Opportunities (30 mins) *Note that the syllabus-in-progress will be left on the board so that ?participants may continue to contribute to it during the poster ?session, final discussion, etc. ORGANIZERS Sumit Basu (sumitb at microsoft dot com) and Ashish Kapoor (akapoor at microsoft dot com) From shivani at MIT.EDU Tue Oct 20 15:43:15 2009 From: shivani at MIT.EDU (Shivani Agarwal) Date: Tue, 20 Oct 2009 15:43:15 -0400 (EDT) Subject: Connectionists: FINAL CFP: NIPS 2009 Workshop - Advances in Ranking Message-ID: ************************************************************************ FINAL CALL FOR PAPERS ---- Advances in Ranking ---- Workshop at the 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009) http://web.mit.edu/shivani/www/Ranking-NIPS-09/ -- Submission Deadline: October 23, 2009 -- ************************************************************************ [ Apologies for multiple postings ] OVERVIEW -------- Ranking problems are increasingly recognized as a new class of statistical learning problems that are distinct from the classical learning problems of classification and regression. Such problems arise in a wide variety of domains: in information retrieval, one wants to rank documents according to relevance to a query; in natural language processing, one wants to rank alternative parses or translations of a sentence; in collaborative filtering, one wants to rank items according to a user's likes and dislikes; in computational biology, one wants to rank genes according to relevance to a disease. Consequently, there has been much interest in ranking in recent years, with a variety of methods being developed and a whole host of new applications being discovered. This workshop aims to bring together researchers interested in the area to share their perspectives, identify persisting challenges as well as opportunities for meaningful dialogue and collaboration, and to discuss possible directions for further advances and applications in the future. One of the primary goals of the workshop will be to reach out to a broad audience. To this end, we will have talks on topics ranging from more statistically/mathematically oriented approaches to ranking, to newer application areas. A second goal will be to bring to the fore a range of questions that are currently being debated within the community, for example via a panel discussion between experts in the field. Overall, the workshop will aim to provide a forum that showcases recent advances in ranking to the broader community, facilitates open debate on some of the questions in this area, and helps catalyze further interest among those new to the topic. FORMAT ------ This is a one-day workshop that will follow the 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009). The workshop will consist of two 3-hour sessions. There will be 2-4 invited talks by experts in the area, and 5-8 short talks. Depending on interest and submissions, we may also have a short poster/demo session. There will be time at the end of each talk/session for questions and discussion. We also intend to have a panel discussion that will be moderated by the organizers and that will bring together leading figures from both academia and industry for discussion and debate. Keynote Lecture --------------- * Persi Diaconis, Stanford University Invited Talks ------------- * Ralf Herbrich, Microsoft Research Cambridge * Lek-Heng Lim, University of California, Berkeley Contributed Talks ----------------- These will be based on submissions of short papers. See below for details. CALL FOR PAPERS --------------- We invite submissions of short papers addressing all aspects of ranking in machine learning, data mining, and statistics, as well as related application areas. These include for example: * algorithms for ranking * mathematical/statistical analyses of ranking * relationships between ranking and other problems * applications of ranking in information retrieval, natural language processing, collaborative filtering, computational biology, computer vision, and other areas * benchmark data sets for ranking * challenges in applying or analyzing ranking methods We also welcome papers on ranking that do not fit into one of the above categories, as well as papers that describe work in progress. Please note that papers that have previously appeared (or have been accepted for publication) in a journal or at a conference or workshop, or that are being submitted to another workshop, are not appropriate for this workshop. All papers presented at the workshop will be made available as electronic proceedings. A print version will be distributed at the workshop. Submission Instructions ----------------------- Submissions should be upto 4 pages in length using NIPS style files (available at http://web.mit.edu/shivani/www/Ranking-NIPS-09/StyleFiles/), and should include the title, authors' names, postal and email addresses, and a brief abstract. Email submissions (in pdf or ps format only) to shivani at mit.edu with subject line "Workshop Paper Submission". The deadline for submissions is Friday October 23, 5:00 pm EDT. Submissions will be reviewed by the program committee and authors will be notified of acceptance/rejection decisions by Wednesday November 11. Final versions of accepted papers will be due on Wednesday November 18. Please note that one author of each accepted paper must be available to present the paper at the workshop. RELATED WORKSHOP ---------------- There is a related workshop at NIPS this year titled Learning with Orderings, with a somewhat different focus than ours; see the webpage above for details. We encourage all participants of our workshop to attend this workshop as well. Submissions should be directed to only one workshop; if a submission cannot be accommodated by one workshop, it may be forwarded to the other workshop for consideration. Please indicate if you would like us to consider this. IMPORTANT DATES --------------- First call for papers -- September 15, 2009 Paper submission deadline -- October 23, 2009 (5:00 pm EDT) Notification of decisions -- November 11, 2009 Final papers due -- November 18, 2009 Workshop -- December 11, 2009 ORGANIZERS ---------- * Shivani Agarwal, MIT * Chris J.C. Burges, Microsoft Research * Koby Crammer, The Technion CONTACT ------- Please direct any questions to shivani at mit.edu. ************************************************************************ From Gunnar.Raetsch at tuebingen.mpg.de Wed Oct 21 03:57:53 2009 From: Gunnar.Raetsch at tuebingen.mpg.de (=?ISO-8859-1?Q?Gunnar_R=E4tsch?=) Date: Wed, 21 Oct 2009 09:57:53 +0200 Subject: Connectionists: Open PhD positions in Tuebingen, Germany Message-ID: <8B82B1E0-FE05-4C1F-B3E8-057F75A94D5E@tuebingen.mpg.de> ###################################################### The * Max Planck research group "Machine Learning in Biology" led by *Gunnar Raetsch* (http://www.fml.mpg.de/raetsch) and the * Interdepartmental Bioinformatics Max Planck research group led by *Karsten Borgwardt* (http://www.kyb.mpg.de/kb) have *openings for several PhD positions in the field of Machine Learning & Computational Biology*. Interested applicants shall apply through the official PhD programme of the Max Planck Institute for Developmental Biology and the Friedrich Miescher Laboratory available at: http://phd.eb.tuebingen.mpg.de Deadline: November 25, 2009. With two Max Planck institutes and the Friedrich Miescher Laboratory, the Max Planck campus in T?bingen offers an ideal environment for interdisciplinary research at the interface of Machine Learning and Biology. The Max Planck Institute for Biological Cybernetics hosts an excellent department for Machine Learning, led by Bernhard Schoelkopf, and the Max Planck Institute for Developmental Biology comprises six departments led by world leaders in their field, including Nobel Prize winner Christiane Nuesslein-Volhard and Leibniz Prize winner Detlef Weigel. ###################################################### -- Dr. Gunnar R?tsch Friedrich Miescher Laboratory, Max Planck Society Spemannstra?e 39, 72076 T?bingen, Germany http://www.fml.mpg.de/raetsch -- Dr. Karsten Borgwardt Interdepartmental Bioinformatics Group Max Planck Institute for Developmental Biology & Max Planck Institute for Biological Cybernetics, T?bingen http://webdav.tuebingen.mpg.de/u/karsten/ From digiovaj at cnel.ufl.edu Wed Oct 21 08:52:42 2009 From: digiovaj at cnel.ufl.edu (Jack Digiovanna) Date: Wed, 21 Oct 2009 14:52:42 +0200 Subject: Connectionists: PhD position at ETH Zurich in vestibular neuroprosthesis project Message-ID: [Introduction] The vestibular system is responsible for the body?s ?sixth sense? ? it detects (along with vision and proprioception) spatial orientation and maintains body equilibrium. If the vestibular system is damaged, patients may experience dizziness, vertigo, or inability to stand up in severe cases. [The CLONS project] The CLONS project is developing a sensory prosthesis for patients with vestibular damage or disorders. It includes both animal and eventually human research. CLONS is funded by the EU under the 'Future and Emerging Technologies Open Scheme'. The PhD student position at ETH Zurich will include modeling, identification of neural dynamics, optimization of stimulation patterns to restore vestibular function, and design of control algorithms. [What we ask] Ideal applicants should have knowledge in one or several of the following subjects: neural engineering, machine learning, neuroscience, signal processing, or biomedical engineering. Experimental skills are also helpful. Applicant must be capable of both independent research and working within a team. Additionally, the applicant should be proficient in English. [About the research partners] Most research will be carried out in the Neuroprosthesis Control Group of the Automatic Control Laboratory at ETH Zurich. Additionally, there will likely be prosthetic testing at the Massachusetts Ear and Eye Infirmary of Harvard Medical School. [To apply] Interested candidates are asked to email their Curriculum Vitae, a list of courses taken and grades obtained, a statement of objectives and research interests (1-2 pages), and contact information of three references to digiovanna at control.ee.ethz.ch. Application deadline is December 4, 2009. Position available from January 2010. ---------------------------------------------- ETH Zurich Jack DiGiovanna, Ph.D. Neuroprosthesis Control Group Automatic Control Laboratory ETL K-24, Physikstrasse 3 8092 Zurich, SWITZERLAND digiovanna at control.ee.ethz.ch control.ee.ethz.ch -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091021/dec4a349/attachment-0001.html From gertito at gmail.com Wed Oct 21 23:34:32 2009 From: gertito at gmail.com (Gert Lanckriet) Date: Wed, 21 Oct 2009 20:34:32 -0700 Subject: Connectionists: Call for contributions: NIPS 2009 Workshop on Understanding Multiple Kernel Learning Methods Message-ID: <4473BB8E-FE0E-46F5-98F0-04ABF8DB1178@gmail.com> ****************************************************************************************************** Call for contributions - Understanding Multiple Kernel Learning Methods Submission deadline: November 3rd, 2009. http://mkl.ucsd.edu/workshop Workshop at the Twenty-Third Annual Conference on Neural Information Processing Systems (NIPS 2009), Whistler, BC, Canada, December 11 or 12, 2009. ****************************************************************************************************** DESCRIPTION Multiple kernel learning has been the subject of nearly a decade of research. Designing and integrating kernels has proven to be an appealing approach to address several, challenging real world applications, involving multiple, heterogeneous data sources in computer vision, bioinformatics, audio processing problems, etc. The goal of this workshop is to step back and evaluate the achievements of multiple kernel learning in the past decade, covering a variety of applications. In short, this workshop seeks to understand where and how kernel learning is relevant (with respect to accuracy, interpretability, feature selection, etc.), rather than exploring the latest optimization techniques and extension formulations. More specifically, the workshop envisions to discuss the following two questions: -- 1 -- Kernel learning vs. kernel design: Does kernel learning offer a practical advantage over the manual design of kernels? -- 2 -- Given a set of kernels, what is the optimal way, if any, to combine them (sums, products, learned or non learned, with or without cross-validation)? We are seeking participants interested in presenting their work and relating their experience in the workshop, providing insight on the above two questions. This includes evidence of MKL improving accuracy beyond any existing method based on single kernels (provided with insights as to why there is such improvement), as well as evidence of the opposite (with insights as to why). We welcome presentation of recent results, as well as presentations based on previously published work that shed light on the above questions. If you are interested in participating and contributing a presentation, please send an email to bmcfee at cs.ucsd.edu with an abstract or a summary, by Tuesday November 3rd, 2009. If the presentation is based on previously published work, please include details of such publications. REPOSITORY In conjunction with the workshop, we are establishing an open repository of data sets for use with MKL algorithms. Authors are encouraged to contribute data to the MKL Repository (mkl.ucsd.edu), and use the repository to benchmark new algorithms. ORGANIZERS * Gert Lanckriet (University of California, San Diego), gert at ece.ucsd.edu * Francis Bach (Ecole Normale Superieure/INRIA), francis.bach at ens.fr * Nathan Srebro (Toyota Technological Institute, Chicago), nati at uchicago.edu * Brian McFee (University of California, San Diego), bmcfee at cs.ucsd.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091022/84a59750/attachment.html From ale at sissa.it Thu Oct 22 11:40:35 2009 From: ale at sissa.it (Alessandro Treves) Date: Thu, 22 Oct 2009 17:40:35 +0200 Subject: Connectionists: Ararat memory meeting Message-ID: <20091022174035.fu7vbt31s8ssskwg@webmail.sissa.it> Looking Back at Mount Ararat: Diversity and cross-Fertilization among Approaches to Memory Onur G?nt?rk?n, Avetis Sadoyan and I invite you to join us on April 5-10, 2010, in Yerevan, Armenia. See http://www.sissa.it/ararat/ The meeting, funded by the Volkswagen Stiftung and supported by EBBS and by PENS, includes a 3-day mini-school, for 40 students, who are invited to apply from Nov 5 on the PENS website (no fees; several travel fellowships will be available) and a 3-day workshop, which we hope many of you will want to attend, to present your memory research. The meeting is free and open, subject to capacity constraints. Accommodation and living expenses will be covered but not, in general, travel costs. Those interested in the workshop, and who share its spirit of enhancing diversity and mutual understanding, are invited to write to me, at ale_treves at yahoo.com, preferably before the end of October (I will be able to respond when returning on e-mail in early November). Please indicate the (approximate) title of a research talk you would contribute. Alessandro -- SISSA - Cognitive Neuroscience, now downtown in via Stock 2/2, V fl BUT NOTE, POSTAL ADDRESS: SISSA, via Beirut 2, 34014 Trieste, Italy tel:39-040-3787623 fax:39-040-3787615 http://people.sissa.it/~ale ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From opossumnano at gmail.com Fri Oct 23 08:01:10 2009 From: opossumnano at gmail.com (Tiziano Zito) Date: Fri, 23 Oct 2009 14:01:10 +0200 Subject: Connectionists: [ANN] Advanced Scientific Programming in Python Winter School in Warsaw, Poland Message-ID: <20091023120109.GC28287@notami.bccn-berlin> Advanced Scientific Programming in Python a Winter School by the G-Node and University of Warsaw Scientists spend more and more time writing, maintaining, and debugging software. While techniques for doing this efficiently have evolved, only few scientists actually use them. As a result, instead of doing their research, they spend far too much time writing deficient code and reinventing the wheel. In this course we will present a selection of advanced programming techniques with theoretical lectures and practical exercises tailored to the needs of a programming scientist. New skills will be tested in a real programming project: we will team up to develop an entertaining scientific computer game. We'll use the Python programming language for the entire course. Python works as a simple programming language for beginners, but more importantly, it also works great in scientific simulations and data analysis. Clean language design and easy extensibility are driving Python to become a standard tool for scientific computing. Some of the most useful open source libraries for scientific computing and visualization will be presented. This winter school is targeted at Post-docs and PhD students from all areas. Substantial proficiency in Python or in another language (e.g. Java, C/C++, MATLAB, Mathematica) is absolutely required. An optional, one-day introduction to Python is offered to participants without prior experience with the language. Date and Location: February 8th ? 12th, 2010. Warsaw, Poland. Preliminary Program: - Day 0 (Mon Feb 8) ? [Optional] Dive into Python - Day 1 (Tue Feb 9) ? Software Carpentry ? Documenting code and using version control ? Test-driven development and unit testing ? Debugging, profiling and benchmarking techniques ? Object-oriented programming, design patterns, and agile programming - Day 2 (Wed Feb 10) ? Scientific Tools for Python ? NumPy, SciPy, Matplotlib ? Data serialization: from pickle to databases ? Programming project in the afternoon - Day 3 (Thu Feb 11) ? The Quest for Speed ? Writing parallel applications in Python ? When parallelization does not help: the starving CPUs problem ? Programming project in the afternoon - Day 4 (Fri Feb 12) ? Practical Software Development ? Software design ? Efficient programming in teams ? Quality Assurance ? Programming project final Applications: Applications should be sent before December 6th, 2009 to: python-winterschool at g-node.org No fee is charged but participants should take care of travel, living, and accommodation expenses. Applications should include full contact information (name, affiliation, email & phone), a *short* CV and a *short* statement addressing the following questions: ? What is your educational background? ? What experience do you have in programming? ? Why do you think ?Advanced Scientific Programming in Python? is an appropriate course for your skill profile? Candidates will be selected on the basis of their profile. Places are limited: early application is recommended. Notifications of acceptance will be sent by December 14th, 2009. Faculty ? Francesc Alted, author of PyTables, Castell? de la Plana, Spain [Day 3] ? Pietro Berkes, Volen Center for Complex Systems, Brandeis University, USA [Day 1] ? Zbigniew J?drzejewski-Szmek, Institute of Experimental Physics, University of Warsaw, Poland [Day 0] ? Eilif Muller, Laboratory of Computational Neuroscience, Ecole Polytechnique F?d?rale de Lausanne, Switzerland [Day 3] ? Bartosz Tele?czuk, Institute for Theoretical Biology, Humboldt-Universit?t zu Berlin, Germany [Day 2] ? Niko Wilbert, Institute for Theoretical Biology, Humboldt-Universit?t zu Berlin, Germany [Day 1] ? Tiziano Zito, Bernstein Center for Computational Neuroscience, Berlin, Germany [Day 4] Organized by Piotr Durka, Joanna and Zbigniew J?drzejewscy-Szmek (Institute of Experimental Physics, University of Warsaw), and Tiziano Zito (German Neuroinformatics Node of the INCF). Website: http://www.g-node.org/python-winterschool Contact: python-winterschool at g-node.org From yann.renard at irisa.fr Fri Oct 23 11:18:13 2009 From: yann.renard at irisa.fr (Yann Renard) Date: Fri, 23 Oct 2009 17:18:13 +0200 Subject: Connectionists: New release of OpenViBE 0.4.0 Message-ID: <4AE1C935.9000502@irisa.fr> New release of *OpenViBE* 0.4.0 is now available for download at : === Overview ========================================= OpenViBE is an opensource platform that enables to design, test and use Brain-Computer Interfaces (BCI). Broadly speaking, OpenViBE can be used in many real-time Neuroscience applications. The OpenViBE platform stands out for its high modularity. It addresses the needs of different types of users (programmers and non-programmers) and proposes a user-friendly graphical language which allows non-programmers to design a BCI without writing a single line of code. OpenViBE is portable, independent of hardware or software targets, can run under Windows and Linux and is entirely based on free and open-source software. OpenViBE is compatible with MATLAB programming. OpenViBE comes with preconfigured scenarios and runs already existing applications such as : * BCI based on motor imagery * P300 speller * Neurofeedback * Real-time visualization of brain activity in 2D or 3D OpenViBE is available under the terms of the LGPL-v2+. The whole software is developed in C++. It consists of a set of software modules that can be integrated easily and efficiently to design BCI applications such as for Virtual Reality interaction. === Where to get more information ==================== If you want more details, check these links : *Website* *Quick introduction video* : *Software download* : *One-hour training session video* : *Screenshots and videos* : === What changed since 0.4.0 RC4 ? =================== In this new release, you will find the following modifiactions (+ for adds, * for modifications, - for removes) : + Mr. Christoph Veigl contributed and added a new driver for OpenEEG Modular EEG / Monolith EEG + We added a new driver : g.Tec's gUSBamp acquisition device + We added a new P300-based entertaining application called "Magic Card" + We added tooltips for new users :) + We updated the sample scenarios * We propose a stabilized P300-based pipeline * We made the development of new classifiers easier thanks to base algorithms * We updated the dependencies installation script for linux so that it uses native packages instead of compiling everything from scratch * We updated the dependencies installation for windows so that DirectX and Visual C++ Runtime gets installed automatically if needed * We updated the online documentation and tutorials * We fixed lots of bugs ! - The VR demo are no more built by default as OpenMASK is not compiling on recent Linux distributions + We added several tooltips for new users :) + We added a k-fold test in the classifier trainer box + We added a functionnality to load/save channel names in the acquisition server + We enabled the voting classifier box to vote either on streamed matrix or on stimulations + We added a frequency band selector box + We added a signal decimation box + We added a CSV file writer box (text based) * We changed the way chanels can be selected in the signal display, power spectrum display and time frequency map display * We reimplemented the common average reference box === What's coming in the next release ================ Here is a snapshot of what we are currently doing and what you can expect from the next release : + A BrainProducts V-Amp acquisition driver + A Neuroscan acquisition driver + A MitsarEEG driver + Online comparison of different processing pipelines performance (e.g. multiple classifiers) + More documentation on the sample scenarios * GUI definition will move from glade to gtk-builder * VR demos will move from OpenMASK to native Ogre3D === Closing words ==================================== We want to thank Christoph Veigl for his quick and efficient contribution. Feel free to join us and to contribute as Christoph and others are doing... ! Also, starting from release 0.4.0, we decided that OpenViBE would be released every three months. You can expect a 0.5.0 release around christmas, 0.6.0 release by the end of march etc... Looking forward to hearing your feedback, we hope you'll enjoy working with OpenViBE as we do. Best regards, The OpenViBE consortium *Contact* : Project Leader : Anatole L?cuyer, INRIA (anatole.lecuyer at irisa.fr) Lead Software Engineer : Yann Renard, INRIA (yann.renard at irisa.fr) From byronyu at stanford.edu Sat Oct 24 04:37:04 2009 From: byronyu at stanford.edu (Byron Yu) Date: Sat, 24 Oct 2009 01:37:04 -0700 (PDT) Subject: Connectionists: Cosyne 2010: Submission Now Open Message-ID: ================================================================= ***** SUBMISSION NOW OPEN ***** ABSTRACT SUBMISSION DEADLINE: 20 Nov 2009 Computational and Systems Neuroscience (Cosyne) MAIN MEETING 25 - 28 Feb, 2010 Salt Lake City, Utah WORKSHOPS 1 - 2 Mar, 2010 Snowbird Ski Resort, Utah http://cosyne.org ================================================================= Cosyne is an annual meeting providing an inclusive forum for the exchange of experimental and theoretical approaches to problems in systems neuroscience. The meeting is expected to draw over 500 researchers from a wide variety of disciplines. The MAIN MEETING is organized in a single track, and consists of both oral and poster sessions. Some oral presentations are invited (see below), while others are selected based on short submitted abstracts. Poster presentations are also selected from the submitted abstracts. The WORKSHOPS are held in 4-8 parallel sessions per day, allowing for more in-depth discussion of specialized topics. CONFIRMED INVITED SPEAKERS: - Keynote: Clay Reid (Harvard Medical School) - Daphne Bavelier (University of Rochester) - Howard Berg (Harvard University) - Adrienne Fairhall (University of Washington) - John Lisman (Brandeis University) - Eve Marder (Brandeis University) - Tirin Moore (Stanford University) - Michael Platt (Duke University) - Nicholas Schiff (Cornell Medical School) - Jackie Schiller (Technion) - Anthony Zador (Cold Spring Harbor Laboratories) Cosyne 2010 will include a special symposium in honour of Horace Barlow, featuring talks by: - Honorary Lecturer: Horace Barlow (Cambridge University) - David Field (Cornell University) - William Geisler (University of Texas) - Geoffrey Hinton (University of Toronto) - Simon Laughlin (Cambridge University) ORGANIZING COMMITTEE: - General Chair: Maneesh Sahani (University College London) - Program Chairs: Anne Churchland (University of Washington) and Bartlett Mel (University of Southern California) - Workshop Chairs: Adam Kohn (Yeshiva University) and Mark Laubach (Yale University) - Communications Chair: Byron Yu (Carnegie Mellon University) EXECUTIVE COMMITTEE: - Anthony Zador (Cold Spring Harbor Laboratory) - Alexandre Pouget (University of Rochester) - Zachary Mainen (Champalimaud Neuroscience Programme) ADVISORY BOARD: - Matteo Carandini (University College London) - Eero Simoncelli (New York University and HHMI) - Peter Dayan (University College London) - Steven Lisberger (UC San Francisco and HHMI) - Karel Svoboda (HHMI Janelia Farm) From rmcantin at isr.ist.utl.pt Sun Oct 25 23:49:03 2009 From: rmcantin at isr.ist.utl.pt (Ruben Martinez-Cantin) Date: Mon, 26 Oct 2009 03:49:03 +0000 Subject: Connectionists: Final CfP: NIPS 09 Workshop on Adaptive Sensing, Active Learning and Experimental Design Message-ID: <24da2db50910252049k20cf0f3apa2d4d5c4e854d75@mail.gmail.com> ---------------------------------------------------------------- ? ? ? ?FINAL?CALL FOR CONTRIBUTIONS ? ? ? ? ? ? ? ? ? NIPS workshop on ?Adaptive Sensing, Active Learning and Experimental Design: ? ? ? ? ? Theory, Methods and Applications ? ? ? Whistler, BC, Canada, December 11, 2009 ? ? http://users.isr.ist.utl.pt/~rmcantin/nips2009.php ---------------------------------------------------------------- Important Dates: ---------------- ? * Submission of extended abstracts: October 27, 2009 ? ? ? (later submission might not be considered for review) ? * Notification of acceptance: November 5, 2009 ? * Workshop date: December 11, 2009 Overview: --------- The fields of active learning, adaptive sensing and sequential experimental design have seen a growing interest over the last decades in a number of communities, ranging from machine learning and statistics to biology and computer vision. Broadly speaking, all active and adaptive approaches focus on closing the loop between data analysis and acquisition. Said in a different way the goal is to use information collected in past samples to adjust and improve the future sampling and learning processes, in the spirit of the twenty questions game. These fields typically address the problem in very diverse ways, and using different problem formulations. The main objective of this workshop is to bring these communities together, share ideas and knowledge, and cross-fertilize the various fields. Invited speakers (confirmed): ----------------------------- ? * Maria-Florina Balcan, Georgia-Tech ? * Donald R. Jones. General Motors Corporation. ? * Andreas Krausse. Caltech ? * Dan Lizotte, University of Michigan ? * Luis Montesano, University of Zaragoza ? * Liam Paninski, Columbia University ? * Matthias Seeger, Saarland University, Saarbruecken Submission instructions: ------------------------ We invite submission of extended abstracts to the workshop. Extended abstracts should beat most 3 pages in length, formatted in according to NIPS style. However, the submission should not be blind. Extended abstracts should be sent in PDF or PS file format by email to alnips09 at gmail.com The selected submission may be accepted either as an oral presentation or as a poster presentation. We encourage participants who can contribute in the following areas: ? * Active learning ? * Active filtering ? * Sequential experimental design ? * Adaptive sensing ? * Optimal information gathering ? * Bayesian optimization ? * Active cognitive development ? * Robotics exploration ? * Sensor placement ? * Active signal processing ? * Online decision making ? * Active model discrimination. ? * Selection criteria/Utility functions ? * Information theoretic metrics The above list is not exhaustive, and we welcome submissions on highly related topics too. Accepted extended abstracts will be made available online at the workshop website. Organizers: ----------- ? * Rui Castro, Columbia University. ? * Nando de Freitas, University of British Columbia. ? * Ruben Martinez-Cantin , Instituto Superior Tecnico. Contact: -------- mailto:alnips09 at gmail.com http://users.isr.ist.utl.pt/~rmcantin/nips2009.php From jason at cs.jhu.edu Mon Oct 26 14:03:44 2009 From: jason at cs.jhu.edu (Jason Eisner) Date: Mon, 26 Oct 2009 14:03:44 -0400 Subject: Connectionists: Call for Summer Team Research Proposals Message-ID: <85bb0a700910261103p437e856albd55d84ea38fd606@mail.gmail.com> 16th Annual JHU Summer Workshop CALL FOR TEAM RESEARCH PROPOSALS Deadline: Wednesday, November 18, 2009. http://www.clsp.jhu.edu/workshops/ws10/CFP The Center for Language and Speech Processing at Johns Hopkins University invites one-page research proposals for a Summer Workshop on Language Engineering, to be held in Baltimore, MD, USA, June 21 to July 30, 2010. Proposals should be suitable for a six-week team exploration, and should aim to advance the state of the art in any of the various fields of Human Language Technology (HLT). This year, proposals in related areas of Machine Intelligence that share techniques with HLT, such as Computer Vision (CV), are also strongly solicited. Proposals are welcome on any topic of interest to HLT, CV and technically related areas. For example, proposals may address novel topics or long-standing problems in one of the following areas. * SPEECH TECHNOLOGY: Proposals are welcomed that address any aspect of information extraction from speech signal (message, speaker identity, language,...). Of particular interest are proposals for techniques whose performance would be minimally degraded by input signal variations, or which require minimal amounts of training data. * NATURAL LANGUAGE PROCESSING: Proposals for knowledge discovery from text are encouraged, as are proposals in traditional fields such as parsing, machine translation, information extraction, sentiment analysis, summarization, and question answering. Proposals may aim to improve the accuracy or enrich the output of such systems, or extend their reach by improving their speed, scalability, and coverage of languages and genres. * VISUAL SCENE INTERPRETATION: New strategies are needed to parse visual scenes or generic (novel) objects, analyzing an image as a set of spatially related components. Such strategies may integrate global top-down knowledge of scene structure (e.g., generative models) with the kind of rich bottom-up, learned image features that have recently become popular for object detection. They will support both learning and efficient search for the best analysis. * UNSUPERVISED AND SEMI-SUPERVISED LEARNING: Novel techniques that do not require extensive quantities of human annotated data to address any of the challenges above could potentially make large strides in machine performance as well as lead to greater robustness to changes in input conditions. Semi-supervised and unsupervised learning techniques with applications to HLT and CV are therefore of considerable interest. Research topics selected for investigation by teams in past workshops may serve as good examples for your proposal (http://www.clsp.jhu.edu/workshops). An independent panel of experts will screen all received proposals for suitability. Results of this screening will be communicated no later than November 20, 2009. Authors passing this initial screening will be invited to Baltimore to present their ideas to a peer-review panel on December 4-6, 2009. It is expected that the proposals will be revised at this meeting to address any outstanding concerns or new ideas. Two or three research topics and the teams to tackle them will be selected for the 2010 workshop. We attempt to bring the best researchers to the workshop to collaboratively pursue the selected topics for six weeks. Authors of successful proposals typically become the team leaders. Each topic brings together a diverse team of researchers and students. The senior participants come from academia, industry and government. Graduate student participants familiar with the field are selected in accordance with their demonstrated performance. Undergraduate participants, selected through a national search, are rising seniors: new to the field and showing outstanding academic promise. If you are interested in participating in the 2010 Summer Workshop we ask that you submit a one-page research proposal for consideration, detailing the problem to be addressed. If your proposal passes the initial screening, we will invite you to join us for the December 4-6 meeting in Baltimore (as our guest) for further discussions aimed at consensus. If a topic in your area of interest is chosen as one of the two or three to be pursued next summer, we expect you to be available for participation in the six-week workshop. We are not asking for an ironclad commitment at this juncture, just a good faith understanding that if a project in your area of interest is chosen, you will actively pursue it. We in turn will make a good faith effort to accommodate any personal/logistical needs to make your six-week participation possible. Proposals should be submitted via e-mail to clsp at jhu.edu by 4PM EST on Wed, November 18, 2009. From auke.ijspeert at epfl.ch Mon Oct 26 09:24:08 2009 From: auke.ijspeert at epfl.ch (Auke Ijspeert) Date: Mon, 26 Oct 2009 14:24:08 +0100 Subject: Connectionists: EPFL Center for Neuroprosthetics / Faculty Positions at the interface of Neuroscience and Bioengineering Message-ID: <4AE5A2F8.5070105@epfl.ch> EPFL Center for Neuroprosthetics / Faculty Positions at the interface of Neuroscience and Bioengineering The Institute of Bioengineering and the Brain-Mind Institute at EPFL invite applications for faculty positions at all ranks, from *tenure track assistant professor to full professor*, for the newly-launched *Center for Neuroprosthetics*. The Center, situated between the School of Engineering and the School of Life Sciences, seeks outstanding individuals working in (1) hearing, and (2) other areas of neuroprosthetics, such as invasive and non-invasive sensing and stimulation in restoration of motor control or sensory perception such as vision. The open faculty positions are offered in an environment of both theoretical and experimental research, rich for the development of novel enabling technologies as well as for seeking deeper understanding of fundamental mechanisms underlying the field of neuroprosthetics. The School of Engineering and the Institute of Bioengineering offer strength in areas that include bio-MEMS/NEMS, bioelectronics, robotics and learning, integrated systems, biomaterials, biophotonics, molecular and computational systems biology, and stem cell biotechnology. The Brain-Mind Institute offers a broader context of neuroscience, with strengths in cognition, behavior, cellular and molecular neuroscience, computational neuroscience, and neurodegeneration, among others. Excellent experimental infrastructure are available including core facilities in animal physiological and behavioral phenomics, animal and human imaging, quantitative light microscopy, genomics and proteomics, micro- and nano-fabrication, and electron microscopy and surface analysis. Successful candidates are expected to initiate independent, creative research programs and participate in undergraduate and graduate teaching. Internationally competitive salaries, start-up resources and benefits are offered. Applications should include a curriculum vitae with a list of publications, a concise statement of research and teaching interests, and the names and addresses (including e-mail) of at least five referees. Applications should be uploaded to: *http://neuroprosthetics-rec.epfl.ch* The deadline for applications is *1 February 2010*. Enquiries may be addressed to: *Prof. Jeffrey A. Hubbell *E-mail: *neuroprosthetics-rec at epfl.ch* For additional information on EPFL, the Schools of Engineering and Life Sciences, the Institute of Bioengineering, and the Brain-Mind Institute, and Institute of Bioengineering, please consult the web sites:* http://www.epfl.ch ,* *http://sti.epfl.ch , http://sv.epfl.ch ,* *http://bmi.epfl.ch *, and *http://ibi.epfl.ch* EPFL aims to increase the presence of women amongst its faculty, and qualified female candidates are strongly encouraged to apply. -- ----------------------------------------------------------------- Prof Auke Jan Ijspeert Associate Professor EPFL-IC-ISIM-GRIJ EPFL, Swiss Federal Institute of Technology, Lausanne Station 14 CH 1015 Lausanne, Switzerland Office: INN 237 Tel: +41 21 693 2658, Fax: +41 21 693 3705 www: http://birg.epfl.ch Email: Auke.Ijspeert at epfl.ch ----------------------------------------------------------------- -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091026/b1f85bc3/attachment-0001.html From kstrong at cambridge.org Mon Oct 26 10:41:44 2009 From: kstrong at cambridge.org (Katy Strong) Date: Mon, 26 Oct 2009 10:41:44 -0400 Subject: Connectionists: New Book in Bayesian Networks from Cambridge University Press Message-ID: New Book from Cambridge University Press! Modeling and Reasoning with Bayesian Networks http://www.cambridge.org/us/catalogue/catalogue.asp?isbn=9780521884389 by Adnan Darwiche Available in Hardback ISBN-13: 9780521884389 Published April 2009 $95.00 "Bayesian networks are as important to AI and machine learning as Boolean circuits are to computer science. Adnan Darwiche is a leading expert in this area and this book provides a superb introduction to both theory and practice, with much useful material not found elsewhere." Stuart Russell, University of California, Berkeley "Bayesian networks have revolutionized AI. This book gives a clear and insightful overview of what we have learnt in 25 years of research, by one of the leading researchers. It is both accessible and deep, making it essential reading for both beginning students and advanced researchers." David Poole, Professor of Computer Science University of British Columbia This book provides a thorough introduction to the formal foundations and practical applications of Bayesian networks. It provides an extensive discussion of techniques for building Bayesian networks that model real-world situations, including techniques for synthesizing models from design, learning models from data, and debugging models using sensitivity analysis. It also treats exact and approximate inference algorithms at both theoretical and practical levels. The author assumes very little background on the covered subjects, supplying in-depth discussions for theoretically inclined readers and enough practical details to provide an algorithmic cookbook for the system developer. More information can be found at: http://www.cambridge.org/us/catalogue/catalogue.asp?isbn=9780521884389 This book will be on display at the 2009 NIPS Conference. For a full list of titles that will be on display, all available at a 20% discount, please visit: http://www.cambridge.org/us/catalogue/promotion.asp?nav=view&code=ME9NIPS Katy Strong Marketing Associate Computer Science, Physics, Mathematics/Statistics Cambridge University Press 32 Avenue of the Americas New York, New York 10013-2473 212.337.6545 p kstrong at cambridge.org To receive email alerts of new Cambridge titles in your field please go to http://www.cambridge.org/alerts From jaakko.peltonen at tkk.fi Tue Oct 27 14:20:02 2009 From: jaakko.peltonen at tkk.fi (jaakko.peltonen@tkk.fi) Date: Tue, 27 Oct 2009 20:20:02 +0200 (EET) Subject: Connectionists: NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics, Extended deadline Message-ID: ------------------------------------------------------------------ FINAL CALL FOR CONTRIBUTIONS - EXTENDED DEADLINE NIPS 2009 workshop on Learning from Multiple Sources with Applications to Robotics Whistler, BC, Canada, December 12, 2009 http://www.dcs.gla.ac.uk/~srogers/lms09/index.htm ------------------------------------------------------------------ Important Dates: ---------------- Submission of extended abstracts: November 2, 2009 (extended deadline) Notification of acceptance: November 10, 2009 Workshop Description: --------------------- Learning from multiple sources denotes the problem of jointly learning from a set of (partially) related learning problems / views / tasks. This general concept underlies several subfields receiving increasing interest from the machine learning community, which differ in terms of the assumptions made about the dependency structure between learning problems. In particular, the concept includes topics such as data fusion, transfer learning, multitask learning, multiview learning, and learning under covariate shift. Several approaches for inferring and exploiting complex relationships between data sources have been presented, including both generative and discriminative approaches. The workshop will provide a unified forum for cutting edge research on learning from multiple sources; the workshop will examine the general concept, theory and methods, and will also examine *robotics* as a natural application domain for learning from multiple sources. The workshop will address methodological challenges in the different subtopics and further interaction between them. The intended audience is researchers working in fields of multi-modal learning, data fusion, and robotics. (More detailed background information is available at the workshop website.) The workshop includes a morning session focused on theory/methods, and an afternoon session focused on the robotics application. The workshop is a core event of the PASCAL2 Network of Excellence. PASCAL2 Invited Speakers: ------------------------- Morning Session: Chris Williams - University of Edinburgh Afternoon Session: Ingmar Posner - University of Oxford Submission Instructions: ------------------------ We invite submission of extended abstracts to the workshop. Extended abstracts should be 2-4 pages, formatted in the NIPS style: http://nips.cc/PaperInformation/StyleFiles Unlike the main NIPS conference, identities of authors do not need to be removed from the extended abstracts. Extended abstracts should be sent in .PDF or .PS file format by email, to either D.Hardoon at cs.ucl.ac.uk or gleen at cis.hut.fi. Acceptance to the workshop will be determined based on peer review of each extended abstract. Submissions are expected to represent high-quality, novel contributions in theory/methods of learning from multiple sources, or high-quality, novel contributions in application of learning from multiple sources to robotics (see below). To encourage participants from the machine learning community to test their algorithms in the domain of robotics, we provide a dataset, with computed features, representative of open research issues in robotics; see the workshop webpage for details. Robotics-oriented papers submitted to the workshop are strongly encouraged to contain an experimental evaluation on the database. The obtained results will be presented by the organizers during the workshop. Submitted extended abstracts may be accepted either as an oral presentation or as a poster presentation; there will be only a limited number of oral presentations in the morning and afternoon sessions. Accepted extended abstracts will be made available online at the workshop website. Depending on the quality of submissions, we will consider preparing a special issue of a journal or a collected volume on the topic of the workshop. A separate call for papers will then be issued after the workshop for the special issue/collected volume. Last year's "Learning from Multiple Sources" workshop led to a special issue in Machine Learning (currently in progress). Organisers ---------- * Barbara Caputo - Idiap Research Institute. * Nicolo Cesa-Bianchi - Universita?degli Studi di Milan. * David Hardoon - Institute for Infocomm Research (I2R). * Gayle Leen - Helsinki University of Technology. * Francesco Orabona - Idiap Research Institure. * Jaakko Peltonen - Helsinki University of Technology. * Simon Rogers - University of Glasgow. Programme Committee ------------------- * Cedric Archambeau - Xerox Research. * Andreas Argyriou - Toyota Technological Institute. * Claudio Gentile - Universita?dell'Insubria. * Mark Girolami - University of Glasgow. * Samuel Kaski - Helsinki University of Technology. * Arto Klami - Helsinki University of Technology. * John Shawe-Taylor - University College London. * Giorgio Valentini - Universita?degli Studi di Milan. Contact Persons --------------- For questions about the workshop, contact David R. Hardoon at D.Hardoon AT cs.ucl.ac.uk. From jclune at msu.edu Tue Oct 27 20:42:22 2009 From: jclune at msu.edu (Jeff Clune) Date: Tue, 27 Oct 2009 20:42:22 -0400 Subject: Connectionists: Generative and Developmental Systems Track, GECCO 2010, Call for Papers and Participation Message-ID: GENERATIVE AND DEVELOPMENTAL SYSTEMS TRACK 2010 GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE (GECCO-2010) http://www.sigevo.org/GECCO-2010 Dear Generative and Developmental Systems (GDS) Researcher, We invite you to submit a paper to the GDS track at GECCO 2010, which we believe is the premier conference for GDS-related work worldwide. Our program committee of GDS experts means that your paper will be reviewed by many leaders in the field. Moreover, the attendance at the GDS track has been high ever since it started in 2007. The size and prestige of the GECCO conference will allow many researchers to learn about your work, both at the conference and via the proceedings (GECCO has the highest impact rating of all conferences in the field of Evolutionary Computation and Artificial Life*). This track focuses on work which has gone by many names, such as artificial development, artificial embryogeny, computational embryology, generative systems, indirect encodings, developmental encodings, genetic regulatory networks (GRNs), Lindenmayer Systems (L-Systems), genotype to phenotype mappings, etc. We invite papers on these topics or related subjects. We also want to alert you to the historical importance of this track. Many of you share our excitement about this field and its connection to the powerful capabilities of biological encoding and development. For many years, there was no consistent annual venue to which researchers in this area could submit papers to be reviewed by a committee of GDS researchers. Most opportunities to present work in this area have been through one- time workshops or symposia that do not carry the same weight as a GECCO conference publication. To address this gap, a committee of GDS researchers encouraged GECCO to establish an official conference track in 2007. With the establishment of the track, we now have an opportunity to build a strong and consistent community that can improve and flourish over time. However, the track can only survive with your submissions. It is important to note that GECCO permanently cancels tracks that fail to attract sufficient submissions. Therefore, our community can preserve this new resource only by contributing papers. We know that you have many options for submitting your GDS-related ideas, and we hope that you will consider the long-term investment that GDS represents for the community. Our program committee is selected from among the top GDS researchers in the world, and we hope that our track will continue to flourish with your enthusiasm. IMPORTANT DATES * Submission deadline: January 13, 2010 * Notification of paper acceptance: March 10, 2010 * Camera-ready submission: April 5, 2010 * GECCO-2010 Conference: July 7-11, 2010 The conference will be held July 7-11, 2010 in the very nice city of Portland, Oregon, USA. For more information, please see the GECCO homepage at http://www.sigevo.org/gecco-2010 Best Regards, Jeff Clune and Julian Miller 2010 GECCO GDS Track Chairs * http://www.cs-conference-ranking.org/conferencerankings/topicsii.html GECCO is sponsored by the Association for Computing Machinery Special Interest Group on Genetic and Evolutionary Computation (SIGEVO). SIG Services: 2 Penn Plaza, Suite 701, New York, NY, 10121, USA, 1-800-342-6626 (USA and Canada) or +212-626-0500 (Global). From rdbeer at indiana.edu Thu Oct 29 09:31:04 2009 From: rdbeer at indiana.edu (Randall Beer) Date: Thu, 29 Oct 2009 09:31:04 -0400 Subject: Connectionists: NSF IGERT Traineeships Available in the Dynamics of Brain-Body-Environment Systems Message-ID: <6BB64378-2207-414D-A671-1D18B570E0E3@indiana.edu> Graduate Traineeships at Indiana University The Cognitive Science Program at Indiana University invites applications from outstanding students for its NSF-funded graduate training program in the dynamics of brain-body-environment systems in behavior and cognition. The goal of the program is to train doctoral students to think across traditional levels of analysis in the cognitive, behavioral and brain sciences. In order to accomplish this goal, we have developed new courses in situated, embodied and dynamical cognitive science, a professional development seminar, summer research internships, an annual research showcase and a colloquium series offering extended opportunities for trainees to interact with visiting speakers. Benefits for students entering this program include a $30,000 annual stipend, full tuition, and coverage of additional fees and health insurance. Our interdisciplinary training group includes cognitive science faculty from the Departments of Psychological and Brain Sciences, Physics, and History & Philosophy of Science, as well as from the School of Informatics and Computing. In addition, we have strong partnerships with top researchers in the fields of dynamical, embodied and situated approaches to behavior and cognition, both nationally and internationally. Applications are due January 15. Only U.S. citizens and permanent residents are eligible for funding through this training program. For more information, contact Dr. Randall Beer at rdbeer at indiana.edu or visit http://igert.cogs.indiana.edu Our program promotes and values a diverse scientific community. -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091029/0331e5d5/attachment.html From maneesh+connectionists at gatsby.ucl.ac.uk Thu Oct 29 10:57:35 2009 From: maneesh+connectionists at gatsby.ucl.ac.uk (maneesh+connectionists@gatsby.ucl.ac.uk) Date: Thu, 29 Oct 2009 14:57:35 +0000 Subject: Connectionists: PhD Study at the Gatsby Unit, UCL Message-ID: Gatsby Computational Neuroscience Unit, UCL 4 year PhD Programme The Gatsby Unit is a centre for theoretical neuroscience and machine learning, focusing on unsupervised, semi-supervised and reinforcement learning, neural dynamics, population coding, Bayesian and nonparametric statistics, kernel methods and applications of these to the analysis of perceptual processing, neural data, natural language processing, machine vision and bioinformatics. It provides a unique opportunity for a critical mass of theoreticians to interact closely with each other, and with other world-class research groups in related departments at UCL (University College London), including Anatomy, Computer Science, Functional Imaging, Physics, Physiology, Psychology, Neurology, Ophthalmology and Statistics, the cross-faculty Centre for Computational Statistics and Machine Learning. We also have links with other UK and overseas universities including Cambridge in the UK, Columbia, New York and the Max Planck Institute in Germany. The Unit always has openings for exceptional PhD candidates. Applicants should have a strong analytical background, a keen interest in machine learning and/or neuroscience and a relevant first degree, for example in Computer Science, Engineering, Mathematics, Neuroscience, Physics, Psychology or Statistics. The PhD programme lasts four years, including a first year of intensive instruction in techniques and research in machine learning and theoretical neuroscience. Competitive fully-funded studentships are available each year (to students of any nationality) and the Unit also welcomes students with pre-secured funding or with other scholarship/studentship applications in progress. Full details of our programme, and how to apply, are available at: http://www.gatsby.ucl.ac.uk/teaching/phd/ For further details of research interests please see: http://www.gatsby.ucl.ac.uk/research.html Applications for 2010 entry (commencing late September 2010) should be received no later than 6th January 2010. Shortlisted applicants will be invited to attend interview in the week commencing 8th March 2010. From m.lengyel at eng.cam.ac.uk Thu Oct 29 12:09:04 2009 From: m.lengyel at eng.cam.ac.uk (=?ISO-8859-1?Q?M=E1t=E9_Lengyel?=) Date: Thu, 29 Oct 2009 16:09:04 +0000 Subject: Connectionists: PhD in Computational Neuroscience at Cambridge University Message-ID: <301A52CF-CC50-4B50-9858-0DF439788F40@eng.cam.ac.uk> Computational Neuroscience @ Cambridge University 3 year PhD Programme The Computational and Biological Learning Lab (CBL) at the Department of Engineering uses engineering approaches to understand the brain and to develop artificial learning systems. Research in computational neuroscience covers learning and memory in perceptual, cognitive, and motor systems. PhD students in the group have the opportunity to pursue computational studies at the neuronal or behavioural level, or experimental studies of human behaviour using state-of-the-art robotic and virtual reality interfaces, or combine computational and experimental approaches. The Department of Engineering has recently received the highest research rating in the UK of all science departments and provides excellent training that includes graduate courses in computational neuroscience and machine learning. CBL is a lively and dynamic group around 30 people, and encourages interaction between all members of the lab, including students, postdocs, and faculty. The entire group meets at least three times a week, on top of various other regular activities, such as reading and journal clubs. Students in computational neuroscience benefit from the strong machine learning group within CBL. Applicants should have * strong problem solving and mathematical skills, * a keen interest in neuroscience, * a relevant first degree, such as Computer Science, Engineering, Mathematics, Neuroscience, Physics, Psychology or Statistics. Students seeking to combine work in neuroscience and machine learning are particularly encouraged to apply. The University has a number of competitive fully-funded studentships available each year (to students of any nationality) and CBL also welcomes students with pre-secured funding or with other scholarship/ studentship applications in progress. Informal enquiries are welcome to Daniel Wolpert (wolpert at eng.cam.ac.uk) or Mate Lengyel (m.lengyel at eng.cam.ac.uk). For more information on CBL see: http://learning.eng.cam.ac.uk/Public/BlgHome For further details of how to apply see: http://learning.eng.cam.ac.uk/Public/Graduate Applications for 2010 entry should be received no later than 15 December 2009. Shortlisted applicants will be invited to attend interview in the week starting with 11 January 2010. -- Mate Lengyel, PhD Computational and Biological Learning Lab Cambridge University Engineering Department Trumpington Street, Cambridge CB2 1PZ, UK tel: +44 (0)1223 748 532, fax: +44 (0)1223 332 662 email: m.lengyel at eng.cam.ac.uk web: www.eng.cam.ac.uk/~m.lengyel From airoldi at fas.harvard.edu Wed Oct 28 19:45:46 2009 From: airoldi at fas.harvard.edu (Edoardo Airoldi) Date: Wed, 28 Oct 2009 19:45:46 -0400 Subject: Connectionists: Final CFP :: NIPS 09 workshop on Graphs & Networks (new deadline) References: <5EC0D4EF-5EB0-4947-9729-F29C68D09F62@fas.harvard.edu> Message-ID: -- Apologies if you receive multiple copies of this announcement -- -- Please forward to anyone who might be interested -- ##################################################################### CALL FOR PAPERS Analyzing Networks and Learning with Graphs a workshop in conjunction with 23nd Annual Conference on Neural Information Processing Systems (NIPS 2009) December 11, 2009, Whistler, BC, Canada http://snap.stanford.edu/nipsgraphs2009/ Deadline for Submissions: Friday, November 4, 2009 **updated** Notification of Decision: Monday, November 9, 2009 ##################################################################### Overview: Recent research in machine learning and statistics has seen the proliferation of computational methods for analyzing networks and learning with graphs. These methods support progress in many application areas, including the social sciences, biology, medicine, neuroscience, physics, finance, and economics. The primary goal of the workshop is to actively promote a concerted effort to address statistical, methodological and computational issues that arise when modeling and analyzing large collection of data that are largely represented as static and/or dynamic graphs. To this end, we aim at bringing together researchers from applied disciplines such as sociology, economics, medicine and biology, together with researchers from more theoretical disciplines such as mathematics and physics, within our community of statisticians and computer scientists. Different communities use diverse ideas and mathematical tools; our goal is to to foster cross-disciplinary collaborations and intellectual exchange. Presentations will include novel graph models, the application of established models to new domains, theoretical and computational issues, limitations of current graph methods and directions for future research. Online Submissions: ------------------- We welcome the following types of papers: 1. Research papers that introduce new models or apply established models to novel domains, 2. Research papers that explore theoretical and computational issues, or 3. Position papers that discuss shortcomings and desiderata of current approaches, or propose new directions for future research. All submissions will be peer-reviewed; exceptional work will be considered for oral presentation. We encourage authors to emphasize the role of learning and its relevance to the application domains at hand. In addition, we hope to identify current successes in the area, and will therefore consider papers that apply previously proposed models to novel domains and data sets. Submissions should be 4-to-8 pages long, and adhere to NIPS format (http://nips.cc/PaperInformation/StyleFiles). Please email your submissions to: nipsgraphs2009 at gmail.com Deadline for Submissions: Friday, November 4, 2009, 11:59 pm (EST) Notification of Decision: Friday, November 6 2009 Workshop Format: ---------------- This is a one-day workshop. The program will feature invited talks, poster sessions, poster spotlights, and a panel discussion. All submissions will be peer-reviewed; exceptional work will be considered for oral presentation. Details about the program will be announced in late November. Invited speakers include: Jennifer Chayes (Microsoft), Matthew Jackson (Stanford), Ravi Kumar (Yahoo), Martina Morris (UW), Cosma Shalizi (CMU). Organizers: ----------- Edo Airoldi, Harvard University Jure Leskovec, Stanford University Jon Kleinberg, Cornell University Josh Tenenbaum, MIT Thank you, we look forward to receiving your submissions! Edo, Jure, Jon & Josh From francois.fleuret at idiap.ch Thu Oct 29 05:03:17 2009 From: francois.fleuret at idiap.ch (Francois Fleuret) Date: Thu, 29 Oct 2009 10:03:17 +0100 Subject: Connectionists: [MASH] Post-doc and phd positions at Idiap, CNRS, WIAS and INRIA Message-ID: <19177.23125.322024.297466@moose.fleuret.org> Dear all, We are starting a European project in machine learning in January 2010 and have several open positions. Best regards, -- Francois Fleuret http://www.idiap.ch/~fleuret/ ---------------------------- snip snip ------------------------------- * ABSTRACT The MASH project is a three-year research initiative which brings together five institutions with expertise in statistics, machine learning, goal planning and computer vision to investigate the collaborative design of complex hand-designed priors for machine learning. MASH is funded by the Information and Communication Technologies division of the European Commission, Cognitive Systems and Robotics unit, under the 7th Research Framework Programme. Research will start in January 2010 and will be carried out in Switzerland (IDIAP), France (CNRS and INRIA), Germany (WIAS) and Czech Republic (CVUT). Open positions are listed below. You can already register on http://www.mash-project.eu to get updated by mail on the status of the project. * SUMMARY The goal of the MASH project is to create new tools for the collaborative development of large families of feature extractors. It aims at starting a new generation of learning software with great prior model complexity. The project is structured around a web platform which will be open to external contributors early in 2010. It comprises collaborative tools such as a wiki-based documentation and a forum, and an experiment center which runs and analyzes experiments on a continuous basis. The applications targeted by the project are classical vision problems, and goal-planning in a 3D video game and with a real robotic arm. Contributors will participate to the project by uploading the source codes of "feature extractors" into the platform. Each one of these extractors processes an input image to generate values relevant to the system. This purposely broad definition spans from classical vision processing such as edge detector or color histogram estimation, to highly dedicated hand-designed templates or event-based memory for the robotic applications. The system concatenates all these extractors to create a very large feature vector, which is used as an input signal for a machine learning algorithm. In practice, anybody can upload such a module at any time. It will be immediately compiled and integrated in the next starting experiment. Preliminary performance measures will be provided in a matter of minutes, and complete results a few hours later. The system encourages contributors to improve upon the work on other and focus on the main weaknesses of the overall system. The scientific issues to be tackled along the course of the project are numerous, from standard machine learning questions such as learning and prediction with very large feature spaces and tight computational constraints, to original problems related to clustering in a functional space. * CONSORTIUM - Idiap Research Institute, Switzerland (IDIAP) - Centre National de la Recherche Scientifique, France (CNRS) - Weierstrass Institute for Applied Analysis and Stochastics, Germany (WIAS) - Institut National de Recherche en Informatique et en Automatique, France (INRIA) - Czech Technical University in Prague, Czech Republic (CVUT) * OPEN PHD POSITION AT IDIAP, SWITZERLAND Contact point: Dr. Fran?ois Fleuret, francois.fleuret at idiap.ch, http://www.idiap.ch/~fleuret/ On-line application at http://www.idiap.ch/~fleuret/hiring-mash.html The selected candidate will be a doctoral student at EPFL EDEE doctoral school. Research will be done at the Idiap Research Institute, under the supervision of Dr. Fran?ois Fleuret. The research to be carried out will be the study of prediction techniques for goal-planning with very large feature space. The candidate will investigate prediction from images, mimicking to learn policies provided by human operators, and extensions of classical Markovian Modeling to the specificity of the MASH project. This work will mix theoretical developments in statistical learning with the implementation of algorithms working on real-world data. Applicants must have a strong background in mathematics and be self-sufficient in programming. They must be familiar with several of the following topics and interested in all of them: probabilities, applied statistics, information theory, signal processing, optimization, algorithmic, and C++ programming. * OPEN POSITIONS AT CNRS, FRANCE Contact point: Dr. Yves Grandvalet, yves.grandvalet at utc.fr, http://www.hds.utc.fr/~grandval/ We have open PhD and PostDoc positions to develop clustering and block-clustering algorithms that will summarize heuristic behaviors across tasks. We aim at providing feedback to the heuristic designers by detecting similar heuristics across similar tasks, thus empowering designers to analyze coexisting strategies, and to detect critical failures. We will develop clustering and block-clustering methods based on probabilistic models and factorization techniques. We will also study the relationships between these approaches. The candidates will hold a Master/PhD in applied mathematics or computer science, and should have interest in both areas. They will work under the supervision of Y. Grandvalet and G. Govaert at the Heudiasyc lab. http://www2.hds.utc.fr/ at University of Technology of Compi?gne http://www.utc.fr/the_university/index.php * OPEN POSITIONS AT WIAS, GERMANY Contact point: Dr. Gilles Blanchard, gilles.blanchard at wias-berlin.de, http://www.wias-berlin.de/people/blanchar/ The research will be carried out at the Weierstrass Institute, Berlin, under the supervision of Dr. G. Blanchard; the selected candidate will be a doctoral student at the Humboldt University, Berlin. The research will concentrate on theoretical and practical developments of prediction techniques from a large set of heterogeneous features: aggregation, sparsification, grouping and reduction techniques, in particular under a strong limitation constraint of the computational burden. Automated construction of a similarity or distance measure between features will be also addressed. Specific Requirements: university degree (at least master/diploma) in mathematics, computer, science or engineering. We expect from potential candidates very good programming skills (C++) and at least basic knowledge in mathematical statistics, theory of machine learning and/or optimization. * OPEN POSITIONS AT INRIA, FRANCE Contact point: Dr. Olivier Teytaud, olivier.teytaud at inria.fr, http://www.lri.fr/~teytaud/ The research will be carried out at the LRI, Universit? Paris-Sud, under the supervision of Olivier Teytaud (INRIA research fellow). We have open PhD and PostDoc positions. The research will focus on theoretical and practical developments of planing techniques from a large set of heterogenous features. Specific Requirements: university degree (at least master) in mathematics, computer science or engineering. We expect from potential candidates very good programming skills (C++) and at least basic knowledge in machine learning and/or planning. ---------------------------- snip snip ------------------------------- From terry at salk.edu Thu Oct 29 22:51:26 2009 From: terry at salk.edu (Terry Sejnowski) Date: Thu, 29 Oct 2009 19:51:26 -0700 Subject: Connectionists: UCSD Computational Neuroscience Graduate Training Program Message-ID: UCSD COMPUTATIONAL NEUROSCIENCE Neurosciences Graduate Training Program - University of California, San Diego http://neurograd.ucsd.edu/doctoral/cnspec.html http://compneuro.salk.edu/ Application deadline: December 1, 2009 http://neurograd.ucsd.edu/admissions/index.html The goal of the Computational Neuroscience Graduate Program at UCSD is to train researchers who are equally at home measuring large-scale brain activity, analyzing the data with advanced computational techniques, and developing new models for brain development and function. Candidates from a wide range of backgrounds are invited to apply, including Biology, Psychology, Computer Science, Physics and Mathematics. The three major themes in the training program are: 1. Neurobiology of Neural Systems: Anatomy, physiology and behavior of systems of neurons. Using modern neuroanatomical, behavioral, neuropharmacological and electrophysiological techniques. Lectures, wet laboratories and computer simulations, as well as research rotations. Major new imaging and recording techniques also will be taught, including two-photon laser scanning microscopy and functional magnetic resonance imaging (fMRI). 2. Algorithms and Realizations for the Analysis of Neuronal Data: New algorithms and techniques for analyzing data obtained from physiological recording, with an emphasis on recordings from large populations of neurons with imaging and multielectrode recording techniques. New methods for the study of co-ordinated activity, such as multi-taper spectral analysis and Independent Component Analysis (ICA). 3. Neuroinformatics, Dynamics and Control of Systems of Neurons: Theoretical aspects of single cell function and emergent properties as many neurons interact among themselves and react to sensory inputs. A synthesis of approaches from mathematics and physical sciences as well as biology will be used to explore the collective properties and nonlinear dynamics of neuronal systems, as well as issues of sensory coding and motor control. Participating Faculty include: * Henry Abarbanel (Physics): Nonlinear and oscillatory dynamics; modeling central pattern generators in the lobster stomatogastric ganglion. * Thomas Albright (Salk Institute): Motion processing in primate visual cortex; linking single neurons to perception; fMRI in awake, behaving monkeys. Director, Sloan Center for Theoretical Neurobiology * Darwin Berg (Neurobiology): Regulation synaptic components, assembly and localization, function and long-term stability. * Ed Callaway (Salk Institute): Neural circuits, visual perception, visual cortex Genetic tools for tracing neural pathways. * Gert Cauwenberghs (Bioengineering/Biology): Neuromorphic Engineering; analog VLSI chips; wireless recording and nanoscale instrumentation for neural systems; large-scale cortical modeling. * Sreekanth Chalasani (Salk): C. elegans: genes, networks and behavior Optical recording of olfactory processing. * Andrea chiba (Cognitive Science): Spatial attention, associative learning, cholinergic neuromodulaiton of behavior, amygdala recordings * EJ Chichilnisky (Salk Institute): Retinal multielectrode recording; neural coding, visual perception. * Garrison Cottrell (Computer Science and Engineering): Dynamical neural network models and learning algorithms * Virginia De Sa (Cognitive Science): Computational basis of perception and learning; multi-sensory integration and contextual influences * Mark Ellisman (Neurosciences, School of Medicine): High resolution electron and light microscopy; anatomical reconstructions. * Fred Gage (Salk Institute): Neurogenesis and models of the hippocampus; neuronal diversity, neural stem cells. * Timothy Gentner (Psychology): Birdsong learning. Neuroethology of vocal communication and audition * Robert Hecht-Nielsen (Electrical and Computer Engineering): Neural computation and the functional organization of the cerebral cortex. * Steve Hillyard (Neurosciences): EEG, perception, attention, memory, Event related potentilas, SSVEP * Harvey Karten (Neurosciences, School of Medicine): Anatomical, physiological and computational studies of the retina and optic tectum of birds and squirrels * David Kleinfeld (Physics): Active sensation in rats; properties of neuronal assemblies; optical imaging of large-scale activity. * William Kristan (Neurobiology): Computational Neuroethology; functional and developmental studies of the leech nervous system, including studies of the bending reflex and locomotion. Director, Neurosciences Graduate Program at UCSD * Herbert Levine (Physics): Nonlinear dynamics and pattern formation in physical and biological systems, including cardiac dynamics and the growth and form of bacterial colonies * Scott Makeig (Institute for Neural Computation): Analysis of cognitive event-related brain dynamics and fMRI using time-frequency and Independent Component Analysis * Javier Movellan (Institute for Neural Computation): Sensory fusion and learning algorithms for continuous stochastic systems * Mikhael Rabinovich (Institute for Nonlinear Science): Dynamical systems analysis of the stomatogastric ganglion of the lobster and the antenna lobe of insects * Pamela Reinagel (Biology): Sensory and neural coding; natural scene statistics; recordings from the visual system of cats and rodents. * John Reynolds (Salk): Visual attention, cortex, psychophysics, neurophysiology, neural modeling * Massimo Scanziani (Biology): Neural circuits in the somotosensory cortex; physiology of synaptic transmission; inhibitory mechanisms. * Terrence Sejnowski (Salk Institute/Neurobiology): Computational models and physiological studies of synaptic, neuronal and network function. * Tanya Sharpee (Salk): Statistical physics and information theory approaches to sensory processing in natural auditory and visual environments. * Gabe Silva (Bioengineering): Cellular neural engineering * Nicholas Spitzer (Neurobiology): Regulation of ionic channels and neurotransmitters in developing neurons and neural function. * Charles Stevens (Salk Institute): Synaptic physiology; theoretical models of neuroanatomical scaling. * Roger Tsien (Chemistry): Second messenger systems in neurons; development of new optical and MRI probes of neuron function, including calcium indicators and caged neurotransmitters * Jing Wang (Biology): Representation of olfactory information in the nervous system of Drosophila * Ruth Williams (Mathematics): Probabilistic analysis of stochastic systems and continuous learning algorithms On-line applications: http://neurograd.ucsd.edu/admissions/index.html The deadline for completed application materials, including letters of recommendation, is December 1, 2008. ----- From ubi at rmki.kfki.hu Fri Oct 30 12:04:36 2009 From: ubi at rmki.kfki.hu (Ujfalussy Balazs) Date: Fri, 30 Oct 2009 17:04:36 +0100 (CET) Subject: Connectionists: BUDAPEST SEMESTER IN COGNITIVE SCIENCE Message-ID: ******************* BUDAPEST SEMESTER IN COGNITIVE SCIENCE (BSCS, www.bscs-us.org), a Hungarian study abroad program, established in 2003, conducted in English for undergraduate students. BSCS offers credit-earning courses for psychology, philosophy, linguistics, biology, computer sciences majors; as well as continuous and optional intensive Hungarian language courses. The programme is complemented by an optional independent research module tailored to students' curricula and research interests. The Programme is hosted by the Department of History and Philosophy of Science of the E?tv?s Lor?nd University (ELTE), Hungary's premium science university established in 1635 and serving as a centre of excellence for modern higher education covering nearly every scientific discipline, with a world-class new campus recently added to its premises where BSCS courses are conducted. Lecturers headlining the program between ?2004-2009 included Professors, in addition to leading academic staff of ELTE, from highly distinguished universities, such as UC Berkeley; Indiana University; University of Sussex; Kalamazoo College, Uinversity Vienna, Technical University Vienna, Hokkaido University, University of Ljubljana etc. For the details click to 'Download program description' at www.bscs-us.org! Send inquiries to P?ter ?rdi (Co-Director, perdi at kzoo.edu) an Szilvia Lehel (BSCS Study Abroad Programme Manager, sylvialehel at yahoo.com). ************* From friedhelm.schwenker at uni-ulm.de Fri Oct 30 13:36:05 2009 From: friedhelm.schwenker at uni-ulm.de (Friedhelm Schwenker) Date: Fri, 30 Oct 2009 18:36:05 +0100 Subject: Connectionists: ANNPR 2010, extended submission deadline Message-ID: <4AEB2405.90901@uni-ulm.de> ================================================================ Call for Papers IAPR Workshop on Artificial Neural Networks for Pattern Recognition (ANNPR 2010) April 11-13, 2010, Nile University, Cairo www.informatik.uni-ulm.de/ni/ANNPR10/index.html ================================================================ ************ new submission deadline : November, 16, 2009 **************************** ANNPR 2010 follows the success of ANNPR 2003 (Florence), ANNPR 2006 (Ulm), and ANNPR 2008 (Paris). This 4th IAPR ANNPR workshop will act as a major forum for international researchers and practitioners working in all areas of neural network based pattern recognition to present and discuss the latest research, results, and ideas in these areas. Papers are solicited dealing with neural networks and pattern recognition which emphasize methodological issues arising in applications. They should be related but not limited to the following topics. * Supervised learning. * Unsupervised learning. * Combination of supervized and unsupervized learning. * Feedforward networks and kernel machines * Recurrent and competitive neural networks. * Hierarchical modular architectures and hybrid systems. * Combination of neural networks and Hidden Markov models. * Multiple classifier systems and ensemble methods. * Image processing and segmentation. * Sensorfusion and multimodal processing. * Feature extraction, dimension reduction. * Clustering and vector quantisation. * Speech and speaker recognition. * Data, text, and web mining. * Bioinformatics. Potential participants should submit a paper describing their work in one of the areas described above. Proceedings will be published as a volume in the Springer LNAI, maximum paper length is 12 pages in LNCS/LNAI format. Instructions for authors, LaTeX templates,etc are available at the (Springer LNCS/LNAI website) . Submission of a paper constitutes a commitment that, if accepted, one or more authors will attend the workshop. Electronic submission in camera-ready format is required. Please submit papers to the ANNPR chairs (Friedhelm Schwenker and Neamat El Gayar) annpr2010 at uni-ulm.de * Important Dates: * Paper submission: November 16, 2009 Notification of acceptance: December 22, 2009 Camera ready copies: January 20, 2010 ==================================================================================== -------------- next part -------------- An HTML attachment was scrubbed... URL: https://mailman.srv.cs.cmu.edu/mailman/private/connectionists/attachments/20091030/f7d9bee8/attachment.html From gianni at idsia.ch Fri Oct 30 14:33:29 2009 From: gianni at idsia.ch (Gianni Di Caro) Date: Fri, 30 Oct 2009 19:33:29 +0100 Subject: Connectionists: CFP: EvoCOMNET 2010, 7th European Workshop on Nature-inspired Techniques for Telecommunications Networks and other Parallel and Distributed Systems Message-ID: <4AEB3179.1080107@idsia.ch> ------------------------------------------------------------------- Call For Papers *** EvoCOMNET 2010 *** 7th European Workshop on Nature-inspired Techniques for Telecommunications Networks and other Parallel and Distributed Systems * Istanbul, Turkey, April 7-9, 2010 * ---------------------------------------------- Extended submission deadline: 11 November 2009 Notification of acceptance: 6 January 2010 Camera ready papers: 15 January 2010 ---------------------------------------------- http://www.evostar.org/ -- Part of the EVOSTAR 2010 events: http://www.evostar.org -- ------------------------------------------------------------------- Many biological systems and processes are characterized by a parallel and distributed architecture in which a large number of autonomous and minimalist units synergistically generate global-level behaviors through local interactions, communications, and the adoption of relatively simple stochastic action policies. The resulting global-level behaviors usually show a number of properties essential for success in natural environments such as: adaptivity to environmental variations, robustness to internal changes and failures, and effectiveness and scalability of performance. Because of all these architectural and performance properties, the observation and reverse-engineering of successful processes in organic, inorganic, and animal systems in nature, has drawn in recent years the attention of many researchers and engineers working in the fields of parallel and distributed systems, and, more in particular, in telecommunications networks. In these domains, nature has provided basic inspiration for the definition of a number of novel algorithms and computational frameworks able to deal effectively with the challenges of current networked systems, which show a growing structural and computational complexity and are made of a large number of highly dynamic and heterogeneous components. The aim of the workshop is to provide a forum to present cutting edge research on nature-inspired approaches to problems arising in the design, control, protection, and management of network systems, and to outline new trends in parallel nature-inspired computation for the solution of complex problems. EvoCOMNET is part of EVOSTAR (EVO*), Europe's premier co-located events in the field of evolutionary and nature-inspired computing. EVO* includes the EuroGP, EvoCOP and EvoBIO conferences and a number of workshops collectively entitled EvoWorkshops. EVO* 2010 is the 12th edition of the event, details and cfps can be found at: http://www.evostar.org --------------------------- SELECTED TOPICS OF INTEREST --------------------------- EvoCOMNET 2010 solicits contributions dealing with the application of ideas from natural processes and systems to the definition, analysis, and development of novel parallel and distributed algorithms, and to the solution of problems of practical and theoretical interest in all domains related to network systems. The scope of the workshop emphasizes the contribution of nature-inspired approaches to the following domains: + Network analysis and design + Routing protocols + Transport protocols + Network protection systems + Load balancing + Quality-of-service provisioning + Mobile ad hoc networks + Sensor networks + Network robotics and sensor-actor networks + Distributed search and computation in P2P networks + Parallel and distributed optimization algorithms + Grid computing + Distributed data mining + Tuning and application of hybrid approaches Particularly welcome are papers reporting: * Applications of nature-inspired techniques to novel problems in the domain of telecommunications networks and parallel and distributed systems * Detailed comparative studies of nature-inspired solutions versus more classical/established techniques * Definition of innovative techniques and/or computational frameworks based on biological systems or processes that have not been considered so far in the literature of nature-inspired systems * Analytical studies of the behavior of the proposed systems * Performance evaluation and visualization of parallel and distributed systems inspired by nature * Real-world implementations * Studies based on real-world data sets * Live demonstrations of algorithm behavior ------------------------------ PUBLICATION DETAILS AND AWARDS ------------------------------ + Conference Proceedings: ---------------------- Accepted papers will be published in a volume of the Springer Lecture Notes in Computer Science (LNCS) together with papers from other workshops of the EVO* conference. + Best Paper Award: ---------------- A Best Paper Award will be given to the author(s) of the paper presented at the workshop that will receive the best evaluation marks from the reviewers and the Session Chairs. -------------------- SUBMISSION PROCEDURE -------------------- Please refer to the http://www.evostar.org website for the submission procedure. The maximum length for a paper is 10 PAGES in LNCS format. Papers will be reviewed by at least three reviewers according to a double blind peer process. --------------- WORKSHOP CHAIRS --------------- + Gianni A. Di Caro IDSIA Lugano, Switzerland gianni AT idsia DOT ch + Muddassar Farooq NUCES Islamabad, Pakistan muddassar DOT farooq AT udo DOT edu + Ernesto Tarantino ICAR-CNR Naples, Italy ernesto DOT tarantino AT na DOT icar DOT cnr DOT it --------------- IMPORTANT DATES --------------- * Submission deadline: 11 November 2009 * Notification of acceptance: 6 January 2010 * Camera ready papers: 15 January 2010 * Events: 7-9 April 2010 -------------------------- WORKSHOP PROGRAM COMMITTEE -------------------------- Ozgur B. Akan (Middle East Technical University, Turkey) Enrique Alba (University of Malaga, Spain) Qing Anyong (National University of Singapore, Singapore) Payman Arabshahi, (University of Washington, USA) Mehmet E. Aydin (University of Bedfordshire, UK) Iacopo Carreras, (CREATE-NET, Italy) Arindam K. Das (University of Washington, USA) Falko Dressler (University of Erlangen, Germany) Frederick Ducatelle (IDSIA, Switzerland) Luca Gambardella (IDSIA, Switzerland) Jin-Kao Hao (University of Angers, France) Malcolm I. Heywood (Dalhousie University, Canada) Byrant Julstrom (St. Cloud State University, USA) Graham Kendall (University of Nottingham, UK) Kenji Leibnitz (Osaka University, Japan) Manuel Lozano-Marquez (University of Granada, Spain) Domenico Maisto (ICAR CNR, Italy) Ronaldo Menezes (Florida Institute of Technology, USA) Martin Middendorf (University of Leipzig, Germany) Roberto Montemanni (IDSIA, Switzerland) Chien-Chung Shen (University of Delaware, USA) Tony White (Carleton University, Canada) Lidia Yamamoto (University of Basel, Switzerland) Franco Zambonelli (University of Modena and Reggio Emilia, Italy) Nur Zincir-Heywood (Dalhousie University, Canada) ------------------------------------------------------------------- From a.k.seth at sussex.ac.uk Sat Oct 31 07:40:55 2009 From: a.k.seth at sussex.ac.uk (Anil Seth) Date: Sat, 31 Oct 2009 11:40:55 +0000 Subject: Connectionists: 3yr postdoc position available in consciousness science Message-ID: <4AEC2247.7030308@sussex.ac.uk> A full-time 3yr post-doctoral position is available within the new multidisciplinary Sackler Centre for Consciousness Science (SCCS) at the University of Sussex, starting early 2010. This research initiative is funded by a founding donation from the Mortimer and Theresa Sackler Foundation. You will work with Dr. Anil Seth (Principal Investigator and SCCS co-director), Prof. Hugo Critchley (Chair in Psychiatry and SCCS co-director) and other researchers in the group, on developing and testing cognitive/computational neuroscience accounts of neural mechanisms underlying consciousness, in health and in disease. The post has a broad remit with opportunities to follow your own research interests within the area of cognitive/computational neuroscience relevant to consciousness. Emphasis will be given to research that integrates functional neuroimaging, behavioural experiments, and computational modelling and analysis. The SCCS has access to multiple neuroimaging methods including fMRI, EEG, and TMS, as well as excellent computational resources. Candidates must have a PhD or equivalent degree in a quantitative science discipline. Prior postdoctoral experience is preferred, as are candidates with a strong background in cognitive neuroscience and neural modelling/analysis. For more information and for how to apply, please see http://www.jobs.ac.uk/job/AAF848/. A second 3yr position with a more clinical focus will shortly be available in the same centre. -- Anil Seth, D.Phil. Reader, EPSRC Leadership Fellow, Dept of Informatics, University of Sussex, Falmer, Brighton, BN1 9QJ, UK W: www.anilseth.com, T: +44 1273 678549,