From mail at jan-peters.net Mon Oct 1 07:59:02 2007 From: mail at jan-peters.net (Jan Peters) Date: Mon, 01 Oct 2007 11:59:02 -0000 Subject: Connectionists: [NIPS 2007 WORKSHOP] Reminder/Call for Posters-Robotics Challenges for Machine Learning Message-ID: *** Apologies for Multiple Postings *** ======== ==== CALL FOR POSTERS ==== =========== NIPS 2007 WORKSHOP: Robotics Challenges for Machine Learning Dates: 7 December, 2007 Organizers: Jan Peters (Max Planck Institute for Biological Cybernetics & USC), Marc Toussaint (Technical University of Berlin) WWW: http://www.robot-learning.de email: nips07 at robot-learning.de Poster request: Submit a one page abstract!!! Abstract Submission Deadline: October 21, 2007 Acceptance Notification: October 26, 2007 ======== ==== CALL FOR POSTERS ==== =========== Abstract: Creating autonomous robots that can assist humans in situations of daily life is a great challenge for machine learning. While this aim has been a long standing vision of robotics, artificial intelligence, and the cognitive sciences, we have yet to achieve the first step of creating robots that can accomplish a multitude of different tasks, triggered by environmental context or higher level instruction. Despite the wide range of machine learning problems encountered in robotics, the main bottleneck towards this goal has been a lack of interaction between the core robotics and the machine learning communities. To date, many roboticists still discard machine learning approaches as generally inapplicable or inferior to classical, hand-crafted solutions. Similarly, machine learning researchers do not yet acknowledge that robotics can play the same role for machine learning which for instance physics had for mathematics: as a major application as well as a driving force for new ideas, algorithms and approaches. Some fundamental problems we encounter in robotics that equally inspire current research directions in Machine Learning are: -- learning and handling models, (e.g., of robots, task or environments) -- learning deep hierarchies or levels of representations (e.g., from sensor & motor representations to task abstractions) -- regression in very high-dimensional spaces for model and policy learning -- finding low-dimensional embeddings of movement as an implicit generative model -- methods for probabilistic inference of task parameters from vision, e.g., 3D geometry of manipulated objects -- the integration of multi-modal information (e.g., proprioceptive, tactile, vision) for state estimation and causal inference -- probabilistic inference in non-linear, non-Gaussian stochastic systems (e.g., for planning as well as optimal or adaptive control) Robotics challenges can inspire and motivate new Machine Learning research as well as being an interesting field of application of standard ML techniques. Inversely, with the current rise of real, physical humanoid robots in robotics research labs around the globe, the need for machine learning in robotics has grown significantly. Only if machine learning can succeed at making robots fully adaptive, it is likely that we will be able to take real robots out of the research labs into real, human inhabited environments. To do so, future robots will need to be able to make proper use of perceptual stimuli such as vision, proprioceptive & tactile feedback and translate these into motor commands. To close this complex loop, machine learning will be needed on various stages ranging from sensory-based action determination over high-level plan generation to motor control on torque level. Among the important problems hidden in these steps are problems which can be understood from the robotics and the machine learning point of view including perceptuo-action coupling, imitation learning, movement decomposition, probabilistic planning problems, motor primitive learning, reinforcement learning, model learning and motor control. Format: The goal of this one-day workshop is to bring together people that are interested in robotics as a source and inspiration for new Machine Learning challenges, or which work on Machine Learning methods as a new approach to robotics challenges. In the robotics context, among the questions which we intend to tackle are Reinforcement Learning, Imitation, and Active Learning: * What methods from reinforcement learning scale into the domain of robotics? * How can we improve our policies acquired through imitation by trial and error? * Can we turn many simple learned demonstrations into proper policies? * Does the knowledge of the cost function of the teacher help the student? * Can statistical methods help for generating actions which actively influencing our perception? E.g., Can these be used to plan visuo-motor sequences that will minimize our uncertainty about the scene? * How can image understanding methods be extended to provide probabilistic scene descriptions suitable for motor planning? Motor Representations and Control: * Can we decompose human demonstrations into elemental movements, e.g., motor primitives, and learn these efficiently? * Is it possible to build libraries of basic movements from demonstration? How to create higher-level structured representations and abstractions based on elemental movements? * Can structured (e.g., hierarchical) temporal stochastic models be used to plan the sequencing and superposition of movement primitives? * Is probabilistic inference the road towards composing complex action sequences from simple demonstrations? Are superpositions of motor primitives and the coupling in timing between these learnable? * How to generate compliant controls for executing complex movement plans which include both superposition and hierarchies of elemental movements? Can we find learned versions of prioritized hierarchical control? * Can we learn how to control in task-space of redundant robots in the presence of under-actuation and complex constraints? Can we learn force or hybrid control in task-space? * Is real-time model learning the way to cope with executing tasks on robots with unmodeled nonlinearities and manipulating uncertain objects in unpredictable environmental interactions? * What new regression techniques can help real-time model learning to improve the execution of tasks on robots with unmodeled nonlinearities and manipulating uncertain objects in unpredictable environmental interactions? Learning structured models and representations: * What kind of probabilistic models provide a compact and suitable description of real-world environments composed of manipulable objects? * How can abstractions or compact representations be learnt from sensori-motor data? * How can we extract features of the sensori-motor data that are relevant for motor control or decision making? E.g., can we extract visual features of objects directly related to their manipulability or ``affordance''? Posters: We are open for any posters posing problems for machine learning and for presenting machine learning algorithms with applications in robotics. Please send us a one page abstract (A4 or letter) describing the poster which you intend to present with at least one reference to previous work. Choose a format of your liking, e.g., the standard NIPS template. The deadline for abstract submissions is October 21, 2007 and the notification will be October 26, 2007 Abstract Submission Deadline: October 21, 2007 Acceptance Notification: October 26, 2007 From steve at cns.bu.edu Mon Oct 1 16:33:24 2007 From: steve at cns.bu.edu (Stephen Grossberg) Date: Mon, 1 Oct 2007 16:33:24 -0400 Subject: Connectionists: autonomous neural system for learning planned action sequences towards a rewarded goal Message-ID: The following article is now available at http://www.cns.bu.edu/Profiles/Grossberg : Gnadt, W. and Grossberg, S. SOVEREIGN: An autonomous neural system for incrementally learning planned action sequences to navigate towards a rewarded goal. Neural Networks, in press. ABSTRACT How do reactive and planned behaviors interact in real time? How are sequences of such behaviors released at appropriate times during autonomous navigation to realize valued goals? Controllers for both animals and mobile robots, or animats, need reactive mechanisms for exploration, and learned plans to reach goal objects once an environment becomes familiar. The SOVEREIGN (Self-Organizing, Vision, Expectation, Recognition, Emotion, Intelligent, Goal-oriented Navigation) animat model embodies these capabilities, and is tested in a 3D virtual reality environment. SOVEREIGN includes several interacting subsystems which model complementary properties of cortical What and Where processing streams and which clarify similarities between mechanisms for navigation and arm movement control. As the animat explores an environment, visual inputs are processed by networks that are sensitive to visual form and motion in the What and Where streams, respectively. Position-invariant and size-invariant recognition categories are learned by real-time incremental learning in the What stream. Estimates of target position relative to the animat are computed in the Where stream, and can activate approach movements toward the target. Motion cues from animat locomotion can elicit head-orienting movements to bring a new target into view. Approach and orienting movements are alternately performed during animat navigation. Cumulative estimates of each movement are derived from interacting proprioceptive and visual cues. Movement sequences are stored within a motor working memory. Sequences of visual categories are stored in a sensory working memory. These working memories trigger learning of sensory and motor sequence categories, or plans, which together control planned movements. Predictively effective chunk combinations are selectively enhanced via reinforcement learning when the animat is rewarded. Selected planning chunks effect a gradual transition from variable reactive exploratory movements to efficient goal-oriented planned movement sequences. Volitional signals gate interactions between model subsystems and the release of overt behaviors. The model can control different motor sequences under different motivational states and learns more efficient sequences to rewarded goals as exploration proceeds. From renaud.jolivet at epfl.ch Tue Oct 2 04:53:58 2007 From: renaud.jolivet at epfl.ch (Renaud Jolivet) Date: Tue, 02 Oct 2007 10:53:58 +0200 Subject: Connectionists: PhD position announcement - High resolution quantitative optical imaging of neurometabolic and neurovascular coupling mechanisms Message-ID: <47020726.4040605@epfl.ch> Noninvasive functional neuroimaging tools are widely applied to study the human brain in action. However, the hemodynamic signals of the most extensively used method are still not well understood. This joint EPFL and ETHZ PhD project aims at understanding the quantitative aspects of hemodynamic changes that occur in the brain. For this aim, novel multimodal high resolution optical imaging tools will be developed and applied in animal experiments. The thesis project (3 years) will consist of the following tasks 1. Development and validation of quantitative spectroscopic optical imaging (Lausanne). 2. In-vivo experiments in the rat and data analysis (Zurich). Applicants should have a strong interest and background in engineering and/or physics. Experience in animal experiments are not mandatory, however, the candidate must be open and interested in this part of the project. Please contact Prof. Dr. Bruno Weber bweber at pharma.unizh.ch or Prof. Dr. Christian Depeursinge christian.depeursinge at epfl.ch. From thomas.shultz at mcgill.ca Mon Oct 1 10:22:29 2007 From: thomas.shultz at mcgill.ca (Thomas Shultz, Dr.) Date: Mon, 1 Oct 2007 10:22:29 -0400 Subject: Connectionists: position at McGill Message-ID: <40EF8CE75AEED54D9A5C542FD3B7AE31056837CB@EXCHANGE2VS3.campus.mcgill.ca> The Psychology Dept. at McGill has a tenure-track position open in quantitative psychology: http://www.psych.mcgill.ca/quan07.htm The search committee will interpret this position rather broadly, so it may be of interest to quantitative modelers. Tom ----------------------------------------------------------------- Thomas Shultz, Professor, Department of Psychology McGill University, 1205 Penfield Ave., Montreal, Quebec, Canada H3A 1B1. Associate Member, School of Computer Science E-mail: thomas.shultz at mcgill.ca Updated 13 September 2007: http://www.psych.mcgill.ca/perpg/fac/shultz/personal/default.htm Phone: 514 398-6139 Fax: 514 398-4896 ----------------------------------------------------------------- From amirhussain007 at aol.com Sun Oct 7 11:24:29 2007 From: amirhussain007 at aol.com (Dr. Amir Hussain) Date: Sun, 7 Oct 2007 16:24:29 +0100 Subject: Connectionists: 1st Call for Papers (CFP): Brain Inspired Cognitive Systems (BICS) 2008, June 24-27, 2008, Brazil In-Reply-To: <47020726.4040605@epfl.ch> Message-ID: <015301c808f6$2876a3e0$0300a8c0@cs.ad.stir.ac.uk> Please circulate the CFP for the third International Brain Inspired Cognitive Systems (BICS 2008) Conference (details below and at: http://www.icsc-naiso.org/conferences/bics2008/bics_2008.html) to interested colleagues and friends.... Many thanks, Dr. Amir Hussain Centre of Cognitive & Computational Neuroscience, University of Stirling, Stirling FK9 4LA, Scotland, UK Email: ahu at cs.stir.ac.uk www.cs.stir.ac.uk/~ahu/ --- 1st Call for Papers (CFP): Brain Inspired Cognitive Systems (BICS) 2008, June 24-27, 2008 http://www.icsc-naiso.org/conferences/bics2008/bics_2008.html General Chair: Allan Kardec Barros Depto. Eng. Eletrica, Universidade Federal do Maranhao to be held at: Pestana S?o Lu?s, S?o Lu?s do Maranh?o Brasil Hotels www.pestana.com/hotels/en/hotels/southamerica/SaoLuisMaranhao/SaoLuis/Home/P estanaSaoLuis.htm Second International ICSC Symposium on Models of Consciousness (MoC 2006) >From foundations to implementations Chair: Ron Chrisley , University of Sussex, U.K. Fourth International ICSC Symposium on Biologically Inspired Systems (BIS 2006) Design and implementation of biologically inspired and neuromorphic systems Chair: Leslie Smith, University of Stirling, U.K. Third International ICSC Symposium on Cognitive Neuro Science (CNS 2006) Models of cognitive systems; Chair: Igor Aleksander, Imperial College London, U.K Fifth International ICSC Symposium on Neural Computation (NC'2006) Progress in neural systems Chair: Amir Hussain, University of Stirling, U.K. Why this conference, and who should attend: The biennial Brain Inspired Cognitive Systems 2008 aims to bring together leading scientists and engineers who use analytic, formal or computational methods both to understand the prodigious processing properties of biological systems, particularly the brain, and to exploit such knowledge to advance technology towards ever higher levels of cognitive competence. The four major symposia are organized in patterns that encourage cross-fertilization across the symposia topics. This emphasizes that, following the success of BICS 2004 (Stirling, Scotland) and BICS 2006 (Greece), BICS 2008 will continue be a major point of contact for researchers and practitioners who can benefit from not only the major advances in their specialist fields but also from the diversity of each other's views. Each of the four mornings is devoted to papers that will be selected for their clear novelty and proven scientific impact, while the afternoons will provide scope for researchers to present their current work and discuss their aims and ambitions. Debates across disciplines will unite researchers with differing perspectives. Deadline for submissions: 31 January 2008 You may submit your abstract at www.x-cd.com/bics08/abstract.cfm SUB-THEMES (including, but not limited to): Models of consciousness: (MoC) Global Workspace Theory Imagination/synthetic phenomenology Virtual Machine Approaches Axiomatic Models Control Theory/Methodology Developmental/Infant Models Will/volition/emotion/affect Philosophical implications Grounding in neurophysiology Enactive approaches Heterophenomenology Cognitive Neuroscience (CNS) Attentional Mechanisms Cognitive Neuroscience of Sensory Modalities CN of volition Affective Systems Language Cortical Models Sub-Cortical Models Cerebellar Models Event location in the brain Others Biologically Inspired Systems (BIS) Brain Inspired (BI) Vision BI Audition and sound processing BI Other sensory modalities BI Motion processing BI Robotics BI Evolutionary systems BI Oscillatory systems BI Signal processing BI Learning Neuromorphic systems Others Neural Computation (NC) NeuroComputational (NC) Hybrid Systems NC Learning NC Control Systems NC Signal Processing Architectures Devices Pattern Classifiers Support Vector Machines Fuzzy or Neuro-Fuzzy Systems Evolutionary Neural Networks Biological Neural Network Models Applications Others Important Dates: July 2008 Acceptance - March/April/2008 Submission deadline - January/2008 Publications: selected, expanded and revised BICS2008 papers will be published in follow-on special issues of international journals ORGANIZED BY: Planning Division ICSC Interdisciplinary Research NAISO Natural and Artificial Intelligence Systems Organization Canada Email: planning at icsc.ab.ca Planning Division ICSC Interdisciplinary Research NAISO Natural and Artificial Intelligence Systems Organization Canada --------------------------------------------------- Email: planning at icsc.ab.ca Website: www.icsc-naiso.org From mseeger at gmail.com Wed Oct 10 12:06:40 2007 From: mseeger at gmail.com (Matthias Seeger) Date: Wed, 10 Oct 2007 18:06:40 +0200 Subject: Connectionists: Call for contributions: NIPS 2007 Workshop on Approximate Bayesian Inference in Continuous and Hybrid Models Message-ID: <43c7cd3f0710100906g461794dfha7dae96b2891078b@mail.gmail.com> Approximate Bayesian Inference in Continuous/Hybrid Models Workshop Neural Information Processing Systems, 7 December, 2007. Whistler, CA ============================================================================== CALL FOR CONTRIBUTIONS: http://intranet.cs.man.ac.uk/ai/nips07 abichm at gmail.com Submission Deadline: October 26, 2007 Sponsored by the Pascal Network of Excellence and Microsoft Research Cambridge ============================================================================== Approximate inference techniques are often fundamental to machine learning successes and this fast moving field has recently facilitated solutions to large scale problems. The workshop will provide a forum to discuss unsolved issues, both practical and theoretical, pertaining to the application of approximate Bayesian inference. The emphasis of the workshop will be in characterizing the complexity of inference and the differential strengths and weaknesses of available approximation techniques. Submissions pertaining to approximate inference in both continuous and discrete variable models are welcome. The target audience are practitioners, providing insight into and analysis of problems with certain methods or comparative studies of several methods, as well as theoreticians interested in characterizing the hardness of inference or proving relevant properties of an established method. ============================================================================= FORMAT: The workshop will be single-day, comprising of a tutorial introduction, invited talks (20 to 30 mins), and presentations of contributed work, with time for discussions. Depending on quality and compatibility with workshop aims, slots for brief talks and posters will be allocated. We intend to have an interactive workshop, and will give priority to contributions of novel ideas not yet established in Machine Learning, and to critical and careful empirical comparative studies over polished applications of established methods to standard problems. We encourage contributions from related fields such as * Statistics (e.g. Markov Chain Monte Carlo methods) * Information Geometry * Filtering, Dynamical Systems * Computer Vision Contributions should be communicated to the program committee (the organizers) in form of an extended abstract (up to 8 pages in the NIPS conference paper style), sent to abichm at gmail.com. ============================================================================= Speakers: David Wipf (UCSF) Manfred Opper (TU Berlin) Jason Johnson (MIT) John Winn (MSR Cambridge) Organizers: Matthias Seeger Max-Planck Biological Cybernetics, Tuebingen David Barber University College London Neil Lawrence University of Manchester Onno Zoeter Microsoft Research Cambridge From P.J.Lisboa at ljmu.ac.uk Wed Oct 3 05:56:11 2007 From: P.J.Lisboa at ljmu.ac.uk (Lisboa, Paulo) Date: Wed, 03 Oct 2007 09:56:11 -0000 Subject: Connectionists: ESANN'08 Special Session CFP: Machine Learning Methods in Cancer Research Message-ID: *** CALL FOR PAPERS *** Apologies for crossposting SPECIAL SESSION: Machine learning methods in cancer research Organizers: Alfredo Vellido (Tech. Univ. Catalunya, Spain), Paulo J.G. Lisboa (Liverpool John Moores University, U.K.) *** As part of ESANN'2008 *** 16th European Symposium on Artificial Neural Networks Advances in Computational Intelligence and Learning Bruges (Belgium) - April 23-24-25, 2008 THE SESSION IN BRIEF: Neural Networks and Machine Learning methods in general are widely used in cancer research and published in clinical, as well as methodological journals. Their acceptance among medical practitioners is steadily increasing, in part because of demands for advanced data analysis relating to bioinformatics, but also because of a realization that decision support will be inherent in the current agenda for personalized medicine. The application of Machine Learning to medical data may be said to have entered a period of adolescence, where the early excitement about their potential has been tempered by the need to assure generality through the use of principled approaches to complexity control. The excitement that was communicated during the early phase of development in the late 90's seems to have whetted the appetite of clinicians for what these methods can achieve, initiating close and fruitful collaborations where key clinical questions are driving new data-based studies, so building clinical relevance, rather than obsolescence, into study design. Machine Learning methods can be applied to a wide range of data types and problems in cancer research. The range of applications includes exploratory analysis and predictive inference, with topics ranging from clustering, through classification, survival analysis, and rule extraction. Hot topics include knowledge discovery from data, but also the integration of multimodal data into clinical inference systems, the use of graphical models for structure finding in large sparse data sets, and methods for robust performance estimation which include the use of automatic rule extraction methods to match inference making with clinical expert knowledge. This special session aims to bring together methodological advances and clinical relevant case studies of Machine Learning approaches to cancer diagnosis and prognosis, and oncology-related bioinformatics. ESANN 2008 participants would benefit from the coming together of a number of internationally renowned experts in the field, who would provide their expert view on a broad palette of state-of-the-art theoretical developments and applications. Further information on the web: www.dice.ucl.ac.be/esann/index.php?pg=specsess#Machine%20learning%20meth ods%20in%20cancer%20research www.lsi.upc.edu/~avellido/research/ESANN08-SpecialSessionCFP.html ***** Deadline for submission of papers: November 23, 2007 ***** The electronic submission procedure is described on the ESANN portal http://www.dice.ucl.ac.be/esann/ CONTACT Alfredo Vellido, PhD Department of Computing Languages and Systems. Polytechnic University of Catalonia Barcelona, Spain Tel.: +34 93 4137796 Fax: +34 93 4137833 email: avellido at lsi.upc.edu Paulo J.G. Lisboa, PhD School of Computing and Mathematical Sciences. Liverpool John Moores University Liverpool, United Kingdom Tel.: +44 151 2312225 Fax: +44 151 2074594 email: P.J.Lisboa at ljmu.ac.uk ESANN'2008 is organized in collaboration with the UCL (Universite catholique de Louvain, Louvain-la-Neuve) and the KULeuven (Katholiek Universiteit Leuven). The conference is technically co-sponsored by the International Neural Networks Society, the European Neural Networks Society, the IEEE Computational Intelligence Society, the IEEE Region 8, the IEEE Benelux Section (sponsors to be confirmed). Proceedings and journal special issue ------------------------------------- The proceedings will include all communications presented to the conference (tutorials, oral and posters), and will be available on-site. Extended versions of selected papers will be published in the Neurocomputing journal (Elsevier). From bressler at fau.edu Thu Oct 11 17:36:06 2007 From: bressler at fau.edu (Steven L Bressler) Date: Thu, 11 Oct 2007 17:36:06 -0400 Subject: Connectionists: PhD Program in Complex Systems and Brain Sciences Message-ID: <009501c80c4e$bb49af90$331e5b83@opal> PH.D. PROGRAM IN COMPLEX SYSTEMS AND BRAIN SCIENCES AT FLORIDA ATLANTIC UNIVERSITY PREDOCTORAL FELLOWSHIPS, RESEARCH & TEACHING ASSISTANTSHIPS The aim of this program is to train scientists who are both mathematically and biologically literate so that they can fully participate in multi-disciplinary research to bring new ways of thinking into neuroscience. Individuals with undergraduate degrees in any pertinent discipline are invited to apply for this 5-year training program at the FAU Center for Complex Systems and Brain Sciences. Graduate training consists of a core curriculum in nonlinear dynamics, neuroscience, computational modeling and cognitive science. Research areas include sensorimotor coordination and learning, human brain imaging, including functional magnetic resonance imaging, EEG, brainstem mechanisms of behavior, neural growth and development, cellulr neurosciences, ion channel dynamics, speech production and perception, neurolinguistics, visual perception, music perception and mathematics of complex systems. Applicants should complete the application package that can be found on our website http://www.ccs.fau.edu and send it together with a letter of interest, vitae and 3 letters of reference to: Rhona Frankel, Center for Complex Systems and Brain Sciences, BS-12, Florida Atlantic University, 777 Glades Road, Boca Raton, FL 33431. E-mail: frankel at ccs.fau.edu. Deadline: January 15th, 2008. Additional mandatory FAU application can be found at http://graduate.fau.edu/GradApp/. --- Steven Bressler, Ph.D. Professor From pavel.laskov at first.fraunhofer.de Mon Oct 15 14:40:45 2007 From: pavel.laskov at first.fraunhofer.de (Pavel Laskov) Date: Mon, 15 Oct 2007 20:40:45 +0200 Subject: Connectionists: NIPS 2007 WORKSHOP: last call for abstracts Message-ID: <4713B42D.4070305@first.fraunhofer.de> *** Apologies for multiple posting *** ========================= CALL FOR ABSTRACTS ========================= NIPS 2007 Workshop on Machine Learning in Adversarial Environments for Computer Security 7 or 8 December, 2007 Whistler, British Columbia, Canada Supported by the PASCAL network of excellence and Google ---------------------------------------------------------------------- Organizers: Pavel Laskov (Fraunhofer Institute FIRST, Germany) Richard Lippmann (MIT Lincoln Laboratory, USA) Deadlines: Extended abstract submission: October 19, 2007 Notification of acceptance: November 2, 2007 Contact: Email: mls-nips07 at first.fraunhofer.de Web page: mls-nips07.first.fraunhofer.de Computer and network security has become an important research area due to the alarming recent increase in hacker activity motivated by profit and both ideological and national conflicts. Increases in spam, botnets, viruses, malware, key loggers, software vulnerabilities, zero-day exploits and other threats contribute to growing concerns about security. In the past few years, many researchers have begun to apply machine learning techniques to these and other security problems. Security, however, is a difficult area because adversaries actively manipulate training data and vary attack techniques to defeat new systems. A main purpose of this workshop is examine adversarial machine learning problems across different security applications to see if there are common problems, effective solutions, and theoretical results to guide future research, and to determine if machine learning can indeed work well in adversarial environments. Another purpose is to initiate a dialog between computer security and machine learning researchers already working on various security applications, and to draw wider attention to computer security problems in the NIPS community. The workshop will consist of invited and contributed presentations as well as panel discussions. Contributions are sought where researchers are applying machine learning to various areas of computer security including but not limited to the following: * Anomaly detection * Intrusion detection * Spam detection * Software vulnerability discovery * Adversary modeling * Adapting to adversarial behavior * Automatic signature generation * Learning adversary behavior * Performance evaluation of adaptive systems * Learning with malicious noise * Hiding and detecting virtual machines * Rootkit detection Submissions should be no longer than two pages and include author contact information and appropriate references to other work. Submissions can contain original contributions as well as summarize prior and recent work. Submissions should be aimed at initiating fruitful discussion of critical issues related to machine learning and computer security, for example by raising controversial issues, sharing open problems, and comparing competing approaches. A limited number of presentations will be given 12 minute oral presentation slots, the remaining accepted submissions will be presented as posters with a 3 minute spotlight. Authors of selected oral presentations are encouraged to present further details as a poster. Substantial time will be reserved for questions and discussion. -- +------------------------------------------------------+ Pavel Laskov, Ph.D. Fraunhofer Institute FIRST IDA Kekulestr. 7, 12489 Berlin tel: +49 30 6392 1870 fax: +49 30 6392 1805 email: pavel.laskov at first.fraunhofer.de University of Tuebingen WSI-RI Sand 13, 72076 Tuebingen tel: +49 7071 29 70574 email: laskov at ri.uni-tuebingen.de http://ida.first.fhg.de/~laskov/ +------------------------------------------------------+ From ale at sissa.it Tue Oct 16 09:31:45 2007 From: ale at sissa.it (Alessandro Treves) Date: Tue, 16 Oct 2007 15:31:45 +0200 Subject: Connectionists: Independent-minded Postdoc in Neural Computation at SISSA, Trieste, Italy Message-ID: <1192541505.4714bd41a4e12@webmail.sissa.it> A postdoctoral position is available from February 1st, 2008, for 3 years, to study the network mechanisms underlying spatial processing and spatial memory in mammals. The research, to be carried out within the LIMBO group at SISSA, is part of a collaboration coordinated by Edvard Moser, and funded by the EU contract SPACEBRAIN. It includes another theoretical group, led by Misha Tsodyks at the Weizmann, and experimental labs in Trondheim, London, Zurich and Heidelberg. At SISSA, activity will focus on mathematical analyses of neural network models, on network simulations and on the analysis of neural activity recorded by the collaborating labs. The ideal candidate is one who brings into the project a perspective different from mine, who is a proficient programmer and a creative thinker. A lack of familiarity with grid cells, the hippocampus and spatial navigation may be advantageous, if combined with an open mind and plastic synapses. The tiny but vibrant Cognitive Neuroscience sector of SISSA includes groups led by Mathew Diamond, Tim Shallice, Raffaella Rumiati and Jacques Mehler and is temporarily located in downtown Trieste. I will review applications from Dec 1st, and until the position is filled. Alessandro Treves -- SISSA - Cognitive Neuroscience, now downtown in via Stock 2/2, V fl BUT NOTE, POSTAL ADDRESS: SISSA, via Beirut 2, 34014 Trieste, Italy tel:972-2-6584945 fax:972-2-6523429 http://people.sissa.it/~ale ---------------------------------------------------------------- SISSA Webmail https://webmail.sissa.it/ Powered by Horde http://www.horde.org/ From erikf at nada.kth.se Tue Oct 16 05:42:40 2007 From: erikf at nada.kth.se (Erik =?iso-8859-1?q?Frans=E9n?=) Date: Tue, 16 Oct 2007 11:42:40 +0200 Subject: Connectionists: Master Program in Computational and Systems Biology Message-ID: <200710161142.40162.erikf@nada.kth.se> Master Program in Computational and Systems Biology The School of Computer Science and Communication of KTH Royal Institute of Technology (Stockholm, Sweden) wishes to bring to your attention a new Master Program in Computational and Systems Biology, starting in the fall of 2008. The Master Program is of two years, in the Bologna format of European Higher education. A web page with more information is at http://www.csc.kth.se/utbildning/program/compsysbio/home Tuition in Swedish higher education is free, for anyone, from any country. Stockholm/Uppsala is a major center of the biotech/pharma industries in Europe, and one of the major hubs in biomedical research, worldwide. KTH has strong traditions in Biotechnology, Theoretical Computer Science (Algorithmic complexity theory) and many other fields, and pursues a vigorous program in Entrepreneurship and Innovation. For any questions or queries, please feel free to get in touch with me, or with any faculty member as listed on the the program web page. http://www.csc.kth.se/utbildning/program/compsysbio/home Erik Frans?n, Vice director of the MSc Program in Computational and Systems Biology Erik Aurell Director of the MSc Program in Computational and Systems Biology From murphyk2 at gmail.com Wed Oct 17 17:15:20 2007 From: murphyk2 at gmail.com (Kevin Murphy) Date: Wed, 17 Oct 2007 14:15:20 -0700 Subject: Connectionists: 2nd CFP - NIPS workshop on statistical models of networks Message-ID: Dear colleagues, I would like to invite you to participate in a workshop on "Statistical models of networks" on Sat 8th of December at NIPS'07 in Whistler, B.C. Due to several requests, we have extended the deadline for abstract submissions until Oct 31st. Please send 1-2 pages in pdf format to murphyk at cs.ubc.ca. More details can be found below and at http://www.cs.ubc.ca/~murphyk/nips07NetworkWorkshop/ Kevin Murphy Lise Getoor Eric Xing Raphael Gottardo The purpose of the workshop is to bring together people from different disciplines - computer science, statistics, biology, physics, social science, etc - to discuss foundational issues in the modeling of network and relational data. In particular, we hope to discuss various open research issues, such as - How to represent graphs at varying levels of abstraction, whose topology is potentially condition-specific and time-varying - How to combine techniques from the graphical model structure learning community with techniques from the statistical network modeling community - How to integrate relational data with other kinds of data (e.g., gene expression or text data) Six stellar invited speakers (Mark Handcock, Peter Hoff, Stephen Fienberg, Stanley Wasserman, Volker Tresp, Jasmine Zhou) have already accepted. However, we are also soliciting abstracts on related work, to broaden the scope of the workshop and improve the exchange of ideas. Please send 2 pages in pdf format to murphyk at cs.ubc.ca by October 31st. This should describe relevant work that you are currently doing or have recently published. Due to time constraints, accepted abstracts can only be presented as a poster and/or a short spotlight. However, all accepted abstracts will also be posted online. From sbasu at media.mit.edu Fri Oct 19 20:05:56 2007 From: sbasu at media.mit.edu (MLSys'07 Workshop Organizers) Date: Fri, 19 Oct 2007 17:05:56 -0700 Subject: Connectionists: [Reminder: Abstracts due 10/31]: Call for Abstracts: NIPS 2007 Workshop on Statistical Learning Techniques for Solving Systems Problems Message-ID: <1192838756.811.1216856457@webmail.messagingengine.com> [Reminder] Call for Abstracts: NIPS 2007 Workshop on Statistical Learning Techniques for Solving Systems Problems (1-2 page abstracts due 10/31) In the last few years, there has been a budding interaction between machine learning and computer systems researchers. In particular, statistical machine learning techniques have found a wide range of successful applications in many core systems areas, from designing computer microarchitectures and analyzing network traffic patterns to managing power consumption in data centers and beyond. However, connecting these two areas has its challenges: while systems problems are replete with mountains of data and hidden variables, complex sets of interacting systems, and other exciting properties, labels can be hard to come by, and the measure of success can be hard to define. Furthermore, systems problems often require much more than high classification accuracy - the answers from the algorithms need to be both justifiable and actionable. Dedicated workshops in systems conferences have emerged (for example, SysML 2006 and SysML 2007) to address this area, though they have had little visibility to the machine learning community. A primary goal of this workshop is thus to expose these new research opportunities in systems areas to machine learning researchers, in the hopes of encouraging deeper and broader synergy between the two communities. During the workshop, through various planned overviews, invited talks, poster sessions, group discussions, and panels, we would like to achieve three objectives. First, we wish to discuss the unique opportunities and challenges that are inherent to this area. Second, we want to discuss and identify "low-hanging fruit" that are be more easily tackled using existing learning techniques. Finally, we will cover how researchers in both areas can make rapid progress on these problems using existing toolboxes for both machine learning and systems. We hope that this workshop will present an opportunity for intensive discussion of existing work in machine learning and systems, as well as inspire a new generation of researchers to become involved in this exciting domain. Call for Abstracts: We are seeking 2-page abstracts about recent work at the intersection of machine learning and systems. We welcome preliminary work and work you may plan to publish at a later conference: we are intentionally not creating proceedings for this workshop so that authors are free to submit work to later venues. However, if there is sufficient interest we will explore the possibility of a special issue of a journal or a book seeded with selected papers from the workshop. Please email your abstracts in PDF format to: mlsys07 at yahoogroups.com by October 31, 2007. We will select abstracts for presentation at the workshop by November 7, 2007. Please send any questions you may have about the workshop to this address as well. We look forward to hearing from you! Sincerely, Archana Ganapathi Sumit Basu Emre Kiciman Fei Sha More information is available at http://radlab.cs.berkeley.edu/MLSys From antonior at neuron.ffclrp.usp.br Mon Oct 8 21:26:12 2007 From: antonior at neuron.ffclrp.usp.br (antonior@neuron.ffclrp.usp.br) Date: Mon, 8 Oct 2007 21:26:12 -0400 Subject: Connectionists: 2nd Latin American School on Computational Neuroscience Message-ID: <380-22007102912612241@M2W031.mail2web.com> Applications are invited for LASCON 2008 SECOND LATIN AMERICAN SCHOOL ON COMPUTATIONAL NEUROSCIENCE Ribeirao Preto, Sao Paulo State, Brazil July 13 - August 1 2008 http://neuron.ffclrp.usp.br/page.php?15 ---------------------------------------------------------------------------- ------------ LASCON aims at introducing advanced undergraduate and graduate students to the use of computational methods for modeling neurons and neural circuits. These methods will be illustrated with the use of programs like neuroConstruct, GENESIS, NEURON, XPPAUT and MATLAB. The faculty is composed of researchers with large experience in computational neuroscience and the use of these programs. The school will last for three weeks, and will consist of theoretical lectures and hands on tutorials given by the lecturers and computational exercises made by the students. Students also will have to work on individual research projects, which they will present orally at the end of the school. The school lectures and tutorials will cover the following topics: 1. Realistic neural modeling; 2. Simplified neural modeling and phase-plane methods of analysis; 3. Spike-train statistics and analysis; 4. Plasticity and learning. Faculty (tentative): David Beeman, University of Colorado, Boulder, CO, USA James Bower, University of Texas, San Antonio, TX, USA Avrama Blackwell, George Mason University, Fairfax, VA, USA Ram?n Huerta, University of California, San Diego, CA, USA Roland K?berle, University of Sao Paulo, Sao Carlos, SP, Brazil Rodrigo Oliveira, George Mason University, Fairfax, VA, USA Reynaldo Pinto, University of Sao Paulo, Sao Paulo, SP, Brazil Sidarta Ribeiro, Edmond and Lily Safra International Institute of Neuroscience of Natal, Natal, RN, Brazil John Rinzel, New York University, New York, NY, USA Arnd Roth, University College, London, UK Volker Steuber, University of Hertfodshire, Hatfield, UK School organizer: Antonio Roque, University of Sao Paulo, Ribeirao Preto, SP, Brazil ---------------------------------------------------------------------------- ------------- Antonio Roque Departamento de Fisica e Matematica FFCLRP, Universidade de Sao Paulo 14040-901 Ribeirao Preto-SP Brazil - Brasil Tels: +55 16 3602-3768 (office); +55 16 3602-3859 (lab) FAX: +55 16 3602-4887 E-mails: antonior at neuron.ffclrp.usp.br antonior at ffclrp.usp.br URL: http://neuron.ffclrp.usp.br -------------------------------------------------------------------- mail2web.com ? What can On Demand Business Solutions do for you? http://link.mail2web.com/Business/SharePoint From turner at gatsby.ucl.ac.uk Tue Oct 9 09:49:17 2007 From: turner at gatsby.ucl.ac.uk (Richard Turner) Date: Tue, 9 Oct 2007 14:49:17 +0100 (BST) Subject: Connectionists: NIPS 2007 workshop, Call for Posters - 'Beyond Simple Cells: Probabilistic models for visual cortical processing' Message-ID: ---------- CALL FOR POSTERS ---------- Beyond Simple Cells: Probabilistic models for visual cortical processing NIPS 2007 workshop, December 7 Organizers: Richard Turner, Pietro Berkes, and Maneesh Sahani (Gatsby Computational Neuroscience Unit, UCL) Web page: http://www.gatsby.ucl.ac.uk/~berkes/docs/NIPS07 -------------------------------------- SUBMISSION DETAILS: This one day workshop (description below) will include a poster session. We encourage contributions from workshop participants, especially those wanting to actively take part in the discussion phases. Authors should submit an extended abstract of max 2 pages to the organizers: Richard Turner: turner at gatsby.ucl.ac.uk Pietro Berkes: berkes at gatsby.ucl.ac.uk The abstracts will also be published on our web site. --------------------------------------- DEADLINES: Abstract submission: Oct, 24 Acceptance Notification: Oct, 26 --------------------------------------- WORKSHOP DESCRIPTION: The goal of the workshop is to assess the success of computational models for cortical processing based on probabilistic models and to suggest new directions for future research. Models for simple cells are now well established; how can the field progress beyond them? We will review important questions in the field through both theoretical and experimental lenses. We have invited a panel of experimentalists to provide the latter with the hope of inspiring new research directions of greater general neuroscientific interest. More precisely, the issues to be discussed include: * Moving beyond a two-layer hierarchy: how can we bridge the gap between objects and Gabors? * What experimental results should we attempt to model? Are current comparisons with physiology at all relevant? * What experimental results would we like to have? * What aspects of visual input are relevant for modeling (eye movements, head movements, color etc.)? How relevant is time? * Is the cortical representation best described by energy models or generative models? * What inference and learning algorithms should we use (beyond MAP: MCMC, CD, VB, score matching)? Modelers will present their current work in short talks. The experimentalists, on the other hand, will form a panel of experts which will drive the discussion, bringing their perspective into play, and constructively expose discrepancies with biology. Through this process, the aim is to attack the main list of questions above, and determine which aspects of the vision problem would especially benefit from a stronger collaboration between modelers and experimentalists. In addition we will have a poster session in which attendees will be able to present state of the art work. --------------------------------------- CONFIRMED SPEAKERS: Michael Black Odelia Schwartz Bruno Olshausen Mike Lewicki CONFIRMED PANEL MEMBERS: Dario Ringach Jozsef Fiser Andreas Tolias From longlifelee at gmail.com Wed Oct 10 11:31:11 2007 From: longlifelee at gmail.com (Soo-Young Lee) Date: Thu, 11 Oct 2007 00:31:11 +0900 Subject: Connectionists: Neuarl Infor Processing - Letters & Reviews, Vol.11, Numbers 7-9 Message-ID: The recent issues of NIP-LR (Neural Information Processing - Letters and Reviews) are online at www.nip-lr.info. Vol. 11, No. 7, July 2007 A Simulated Annealing PolyClonal Selection Algorithm and Its Application to Traveling Salesman Problems Shangce Gao, Zheng Tang, Hongwei Dai and Gang Yang Robot Task Learning based on Reinforcement Learning in Virtual Space Tadashi Tsubone, Koichi Sugiyama, and Yasuhiro Wada Vol.11, No.8, August 2007 An Efficient Parallel Algorithm for Maximum Cut Problem Yanqiu Che, Zheng Tang and Shaozhi Liu A Novel Learning Method for Elman Neural Network Using Local Search ZhiQiang Zhang, Zheng Tang and Catherine Vairappan Vol.11, No.9, September 2007 Absolute Stability Analysis for a Class of Discrete-Time Neural Networks Cong Jin Matching Parameter Optimization in Self-Organizing Relationship (SOR) Network by Employing Energy Functions Hideaki Misawa and Takeshi Yamakawa -- Soo-Young Lee Director, Brain Science Research Center, KAIST From ps629 at columbia.edu Fri Oct 5 12:36:54 2007 From: ps629 at columbia.edu (Paul Sajda) Date: Fri, 5 Oct 2007 12:36:54 -0400 Subject: Connectionists: Faculty Position in Neural Engineering Message-ID: <4884A127-6A2D-4F89-9C56-AD98BF439967@columbia.edu> Faculty Position in Neural Engineering Description & Requirement The Department of Biomedical Engineering in the Fu Foundation School of Engineering and Applied Science at Columbia University invites applications for a tenure-track faculty position at the Assistant, Associate or Full Professor level in Neural Engineering. Level of appointment depends on qualification. Specific areas of interest include: neuroimaging, neural tissue engineering, neuromorphic engineering, computational neural modeling and brain machine interfaces. Successful candidates must demonstrate an ability to develop a world-class research program, be capable of obtaining competitive external research funding, and participate in and be committed to outstanding teaching at both the undergraduate and graduate levels. The candidate should have a doctorate in Biomedical Engineering or a related discipline. Application Applicants should send a complete curriculum vitae, three publication reprints, a statement of research interests, a statement of teaching experience and philosophy, and names and contact information for four references to: Professor Paul Sajda, Chair of the Faculty Search Committee, 351 Engineering Terrace, MC 8904, 1210 Amsterdam Avenue, Columbia University, New York, N.Y. 10027 by February 1, 2008. Materials can also be emailed to ps629 at columbia.edu. The search will remain open until the position has been filled. Columbia University is an affirmative action/equal opportunity employer. Women and minorities are encouraged to apply. Paul Sajda, Ph.D. Associate Professor Department of Biomedical Engineering Columbia University 351 Engineering Terrace Building, Mail Code 8904 1210 Amsterdam Avenue New York, NY 10027 tel: (212) 854-5279 fax: (212) 854-8725 email: ps629 at columbia.edu http://liinc.bme.columbia.edu From arthur at tuebingen.mpg.de Wed Oct 10 21:34:22 2007 From: arthur at tuebingen.mpg.de (Arthur Gretton) Date: Thu, 11 Oct 2007 03:34:22 +0200 Subject: Connectionists: Call for papers: NIPS 2007 Workshop on Representations and Inference on Probability Distributions Message-ID: <1192066462.5149.119.camel@localhost> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Call for Papers Workshop on Representations and Inference on Probability Distributions http://nips2007.kyb.tuebingen.mpg.de/ Event: NIPS 2007 Workshop Location: Whistler, Canada Date: 7-8 December 2007 Submission deadline: 01 November Accept/Reject notification: 10 November Supported by the PASCAL network of excellence %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Workshop Overview ================= When dealing with distributions it is in general infeasible to estimate them explicitly in high dimensional settings, since the associated learning rates can be arbitrarily slow. On the other hand, a great variety of applications in machine learning and computer science require distribution estimation and/or comparison. Examples include testing for homogeneity (the "two-sample problem"), independence, and conditional independence, where the last two can be used to infer causality; data set squashing / data sketching / data anonymisation; domain adaptation (the transfer of knowledge learned on one domain to solving problems on another, related domain) and the related problem of covariate shift; message passing in graphical models (EP and associated algorithms); compressed sensing; and links between divergence measures and loss functions. The purpose of this workshop is to bring together statisticians, machine learning researchers, and computer scientists working on representations of distributions for various inference and testing problems, to discuss the compromises necessary in obtaining useful results from finite data. In particular, what are the capabilities and weaknesses of different distribution estimates and comparison strategies, and what negative results apply? Speakers: ========= * Yasemin Altun, MPI for Biological Cybernetics * Tugkan Batu, LSE * Shai Ben-David, University of Waterloo * Gerard Biau, Paris VI University * Carlos Guestrin, CMU * Zaid Harchaoui, ENST * Piotr Indyk, MIT * John Shawe-Taylor, UCL Additional speakers (including submitted talks) will be listed on the website once confirmed. Call for Contributions ====================== We invite submissions for 15 minute contributed talks or posters. The intended emphasis is on recent innovations, work in progress, and promising directions or problems for new research. If you would like to submit to this session, please send an abstract to Arthur Gretton (arthur at tuebingen.mpg.de) by November 01, indicating your preference for a poster or a talk. Please do not send posters or long documents. Decisions as to which proposals are accepted will be sent out on November 10. Workshop Organisers =================== Kenji Fukumizu http://www.ism.ac.jp/~fukumizu Arthur Gretton http://www.kyb.mpg.de/~arthur Alex Smola http://sml.nicta.com.au/~smola/ From oschwart at aecom.yu.edu Fri Oct 12 11:07:52 2007 From: oschwart at aecom.yu.edu (Odelia Schwartz) Date: Fri, 12 Oct 2007 11:07:52 -0400 (EDT) Subject: Connectionists: Postdoctoral position in computational/theoretical neuroscience Message-ID: Postdoctoral position in computational/theoretical neuroscience Applications are invited for a postdoctoral position in computational/theoretical neuroscience, in the lab of Odelia Schwartz at the Albert Einstein College of Medicine in New York City. Our lab employs tools of computational and theoretical neuroscience, to study systems from the neural level through to perception and behavior. For example, we develop models of sensory processing based on the hypothesis that images and sounds have predictable and quantifiable statistical regularities to which the brain is sensitive. The models are constructed through interplay with physiological and psychophysical data. Example projects: (1) Models of how neurons and percepts are affected by contextual information: spatially, what surrounds a given feature or object; and temporally, what has been observed in the past, i.e., adaptation. (2) Models of hierarchical neural processing. (3) Characterizing the statistics of images for classes of scenes and textures, and understanding how manipulating such statistics affect neural processing. For more information about the lab and recent publications, see: http://neuroscience.aecom.yu.edu/faculty/primary_faculty_pages/schwartz.html The candidate should have a Ph.D. in a relevant discipline, a strong quantitative background, and an interest in neuroscience. Prior experience would ideally include areas such as computational neuroscience, machine learning, statistics and/or signal processing. Albert Einstein College of Medicine (AECOM) offers a vibrant interdisciplinary environment, with a growing systems and computational contingent. The position will be in the Department of Neuroscience and Center for Bioinformatics, with opportunities for experimental interactions and collaborations. AECOM is located in a quiet neighborhood of New York, only a short subway ride from Manhattan. Information about working at the AECOM, including housing for postdocs, can be found at: http://www.aecom.yu.edu/home/belfer_institute/ Initial appointments are for one year and renewable for a total period of three years. Salary is competitive and will commensurate with experience. Please send inquiries; or a CV, short statement of research interests, and names and contact information of 3 references to: Odelia Schwartz oschwart at aecom.yu.edu Meetings: I will be available to meet at the upcoming SFN and NIPS conferences. From alexander.gepperth at honda-ri.de Mon Oct 15 09:45:58 2007 From: alexander.gepperth at honda-ri.de (Alexander Gepperth) Date: Mon, 15 Oct 2007 15:45:58 +0200 Subject: Connectionists: Large real-world color object dataset available Message-ID: <47136F16.5080705@honda-ri.de> Sorry for multiple postings!! A large dataset of real-world color objects is now available. It consists of the object classes "car" (~1800 examples), "signal board"(~1000 examples) and "clutter"(~3500 examples). The objects were taken from the segmentation results of a vision system operating within a moving car and are thus biased as well as imperfectly segmented, posing challenges to object recognition. The data is partitioned into 4 classes of (heuristically determined) increasing complexity and difficulty. The dataset consists of images in the standard PPM format and can be obtained from http://www.gepperth.eu/alexander/carDatasetsICANN.html The paper in which this dataset is described and used is A. Gepperth, B. Mersch, J. Fritsch, C.Goerick. Color object recognition in real-world scenes. In J.M. de Sa, editor, /ICANN 2007, part II/, Lecture notes in Computer Science, number 4469. Springer Verlag Berlin, Heidelberg, New York, 2007. available from the same site: http://www.gepperth.eu/alexander/papers/gepperthEtAl2007.ps From D.Hardoon at cs.ucl.ac.uk Thu Oct 4 18:22:45 2007 From: D.Hardoon at cs.ucl.ac.uk (David R. Hardoon) Date: Thu, 04 Oct 2007 22:22:45 -0000 Subject: Connectionists: 'Music, Brain & Cognition' Workshop announcement Message-ID: <5D607E64-F760-4D78-B0C4-633B88D45D57@cs.ucl.ac.uk> Deadline for submission extended to the 12'th of October. Apologies for cross posting and please forward to whom ever this may be of interest to. NIPS'07 Workshop - Whistler, BC, December 7-8, 2007 Music, Brain & Cognition Day 1: "Learning the Structure of Music and its Effects on the Brain" Day 2: "Models of Sound and Music Cognition" ====================================================== http://homepage.mac.com/davidrh/MBCworkshop07/ Call for contributions ------------ We call for paper contribution of up to 8 pages to the workshop using NIPS style. The accepted papers will be available for downloading from this site. Selected papers will be considered for publication in a special issue of ?Journal of New Music Research?. Day 1 - Machine learning based models for learning the structure of music - Models for predicting style of performers - Analysis and models of fMRI/EEG/MEG scans from musical stimuli (as opposed tosimplistic auditory stimuli) - Predicting music generated patterns in fMRI/EEG/MEG - Strategies for embedding representations of musical experience into generative music / performance systems - Methods for generative musical performance and composition - Generative music and/or performance systems based on models of brain functioning - Similar and further models for learning and analysing the structure of music Day 2 - Computational models of cognitively inspired sound processing - Top down control of musical processing of pitch, onset, timbre - Models of musical memory, saliency, attention - Models of music development and learning - Computer aided sound design - Models as above, applied to other domains (e.g. speech and vision) with potential application in music Accepted papers will be either presented as a talk or poster (with poster spotlight) Papers should be submitted to the organisers D.Hardoon at cs.ucl.ac.uk, hpurwins at iua.upf.edu and please indicate if you wish to present on day 1 or day 2 and whether you only wish to present a poster. Important Dates ------------ Deadline for submissions: October 12, 2007 Notification of acceptance: October 31, 2007 Workshop taking place: December 7-8, 2007 Description ------------ Music is one of the most widespread of human cultural activities, existing in some form in all cultures throughout the world. The definition of music as organised sound is widely accepted today but a na?ve interpretation of this definition may suggest the notion that music exists widely in the animal kingdom, from the rasping of crickets' legs to the songs of the nightingale. However, only in the case of humans does music appear to be surplus to any obvious biological purpose, while at the same time being a strongly learned phenomenon and involving significant higher order cognitive processing rather than eliciting simple hardwired responses. A two day workshop will take place at NIPS 07 (Vancouver, Canada) and will span topics from signal processing and musical structure to the cognition of music and sound. In the first day the workshop will provide a forum for cutting edge research addressing the fundamental challenges of modeling the structure of music and analysing its effect on the brain. It will also provide a venue for interaction between the machine learning and the neuroscience/brain imaging communities to discuss the broader questions related to modeling the dynamics of brain activity. During the second day the workshop will focus on the modeling of sound, music perception and cognition. These have provide, with the crucial role of machine learning, a break-through in various areas of music technology, in particular: Music Information Retrieval (MIR), expressive music synthesis, interactive music making, and sound design. Understanding of music cognition in its implied top-down processes can help to decide which of the many descriptors in MIR are crucial for the musical experience and which are irrelevant. The organisers of the workshop are investigators for three main European projects; Learning the Structure of Music (Le StruM), Closing the loop of Evaluation and Design (CLOSED), Emergent Cognition through Active Perception (EmCAP) Mention. The target group is of researchers within the fields of (Music) Cognition, Music Technology, Machine Learning, Psychology, Sound Design, Signal Processing and Brain Imaging. For more details, please go to http://homepage.mac.com/davidrh/ MBCworkshop07/ Day 1 - http://homepage.mac.com/davidrh/MBCworkshop07/Day_1.html Day 2 - http://homepage.mac.com/davidrh/MBCworkshop07/Day_2.html Organisers ------------ Day 1 * David R. Hardoon (University College London) * Eduardo Reck-Miranda (University of Plymouth) * John Shawe-Taylor (University College London) Day 2 * Hendrik Purwins (Universitat Pompeu Fabra) * Xavier Serra (Universitat Pompeu Fabra) * Klaus Obermayer (Berlin University of Technology) ---------------------------------------------------------------------- "Who dares... wins" Dr. David R. Hardoon The Centre for Computational Statistics & Machine Learning Intelligent Systems Research Group Dept. of Computer Science University College, London Gower Street London, UK WC1E 6BT Tel: +44 (0) 20 7679 0425 Fax: +44 (0) 20 7387 1397 Email: D.Hardoon at cs.ucl.ac.uk www: http://homepage.mac.com/davidrh/ From pascal.fua at epfl.ch Mon Oct 8 16:20:29 2007 From: pascal.fua at epfl.ch (Pascal Fua) Date: Mon, 08 Oct 2007 22:20:29 +0200 Subject: Connectionists: Faculty Positions in Computer Science at EPFL Message-ID: <470A910D.2070504@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, bioinformatics, graphics, machine learning, and design methodologies for integrated systems. Significant start-up resources and research infrastructure will be available. Internationally competitive salaries and benefits are offered. 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 1, 2008. 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 -- -------------------------------------------------------------------- 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 paul.cisek at umontreal.ca Thu Oct 11 12:16:06 2007 From: paul.cisek at umontreal.ca (Paul Cisek) Date: Thu, 11 Oct 2007 12:16:06 -0400 Subject: Connectionists: Postdoctoral position in decision-making at the University of Montreal Message-ID: <007301c80c22$07029410$f2e4cc84@Homunculus> The department of physiology at the University of Montr?al invites applications for a postdoctoral fellowship in neurophysiological and computational studies of decision-making and voluntary behavior control. The successful applicant will possess a PhD in neuroscience or related field. Experience with neurophysiological techniques or computational modeling is highly desirable. This individual will take part in a research project studying the cerebral cortical mechanisms of decision-making in humans and non-human primates using a combination of computational modeling techniques and multi-electrode recording, 3T functional MRI, and/or transcranial magnetic stimulation. Depending on the applicant's qualifications and interests, they will help to design and conduct behavioral and neurophysiological experiments, analyze data, develop theoretical models of neural systems, prepare manuscripts for publication, and participate in international conferences. Starting date of the appointment is flexible, and salary will be determined according to Canadian Institutes of Health Research guidelines. For further information, please contact Dr. Paul Cisek (paul.cisek at umontreal.ca). Applicants are asked to submit a curriculum vita, a statement of research interests, and the names and contact information of three references, to: Dr. Paul Cisek Department of physiology University of Montr?al C.P. 6128 Succursale Centre-ville Montr?al, QC H3C 3J7, CANADA Phone : 514-343-6111 x4355 Web : www.cisek.org/pavel email: paul.cisek at umontreal.ca Applications will be accepted until the position is filled. A preliminary interview at the Society for Neuroscience meeting is possible. ----------------------------------------------- Paul Cisek, Ph.D. Department of physiology, room 4141 University of Montreal C.P. 6128 Succursale Centre-ville Montreal QC H3C 3J7 Canada phone: 514-343-6111 x4355 FAX: 514-343-2111 email: paul.cisek at umontreal.ca ----------------------------------------------- From esann at dice.ucl.ac.be Sat Oct 13 08:03:13 2007 From: esann at dice.ucl.ac.be (esann@dice.ucl.ac.be) Date: Sat, 13 Oct 2007 14:03:13 +0200 Subject: Connectionists: CFP: ESANN'2008 special sessions Message-ID: <005c01c80d91$0859c8e0$43ed6882@maxwell.local> ESANN'2008 16th European Symposium on Artificial Neural Networks Advances in Computational Intelligence and Learning Bruges (Belgium) - April 23-24-25, 2008 Special sessions ============================================= The following message contains a summary of all special sessions that will be organized during the ESANN'2008 conference. Authors are invited to submit their contributions to one of these sessions or to a regular session, according to the guidelines found on the web pages of the conference http://www.dice.ucl.ac.be/esann/. Deadline for submissions: November 23, 2007. According to our policy to limit the number of unsolicited e-mails, we gathered all special session descriptions in a single message, and try to avoid sending it to overlapping distribution lists. We apologize if you receive multiple copies of this e-mail despite our precautions. Special sessions that will be organized during the ESANN'2008 conference ========================================================= 1. Computational Intelligence in Computer Games Colin Fyfe (University of Paisley, United Kingdom) 2. Methodology and standards for data analysis with machine learning tools. Damien Fran?ois (Universit? catholique de Louvain, Belgium) 3. Neural Networks for Computational Neuroscience David Meunier, H?l?ne Paugam-Moisy (LIRIS-CNRS, France) 4. Machine learning methods in cancer research Alfredo Vellido (Polytechnic University of Catalonia, Spain), Paulo J.G.Lisboa (Liverpool John Moores University, United Kingdom) 5. Machine Learning Approches and Pattern Recognition for Spectral Data Thomas Villmann (Univ. Leipzig, Germany), Erzs?bet Mer?nyi (Rice University, USA), Udo Seiffert (IPK Gatersleben, Germany) Short descriptions ================== 1. Computational Intelligence in Computer Games ----------------------------------------------------------------------- Organized by: Colin Fyfe (University of Paisley, United Kingdom) This Special Session of ESANN2008 invites papers on any application of computational intelligence to computer games. The computational intelligence techniques may be artificial neural networks, evolutionary algorithms, artificial immune systems, swarm intelligence or machine learning techniques. The computer games may be any current kind of games (First Person Shooter, Real Time Strategy, driving, simulator, board, puzzle, classic, arcade games...) running on any platform (PC, Mac, Java, Flash, XBOX 360, Playstation 3, Wii...) or computer simulations of classical mathematical game theory problems. In all cases, the paper should demonstrate that the technique used has provided a degree of intelligence in the computer game. 2. Methodology and standards for data analysis with machine learning tools. ----------------------------------------------------------------------- Organized by: Damien Fran?ois (Universit? catholique de Louvain, Belgium) Compared to well established fields, such as linear identification, data mining might still be considered as an art. There is no standard procedure leading from data to models; each practitioner develops its own methodology, depending on the tools he uses and the data he faces. While literature describing the tools is vast, papers explaining how to use them efficiently are rare. The general methodology for data analysis requires making choices as experimentation time is often limited; the choice of the feature selection method, the choice of the model family (e.g. ANN, trees, SVM), the choice of the algorithm for choosing the model structure (LOO, CV, bootstrap), the choice of the training algorithm and its parameters (e.g. 3 stage learning vs orthogonal least squares for RBFN), the choice of the splitting of the data in training, validation and test sets, are some examples. Hundreds of methods have been available since decades, but still nobody knows which one to choose. This session is dedicated to methodology-oriented papers describing best practices, or proposing guidelines, that would help data miners in making relevant choices thorough the whole process of data mining. Papers should not propose a new method but rather propose a methodology to choose, or combine existing models/algorithms/methods that are appropriate to solve a given data analysis problem. A non-exhaustive list of questions that could be addressed is the following: - What are the adequate early steps in data exploration for modelling? - Which models/implementations should I have in my data analysis toolbox? - How do I choose a feature selection and/or model selection method? - How do I choose proper preprocessing? How do I treat missing data ? How do I handle non standard data ? - What should be done when classes are unbalanced? - How to deal with heterogeneous features? How do I standardize/normalise them? - When are trees more appropriate than support vector machines or artificial neural networks? - Which datasets should I use to benchmark my new algorithm/model? How do I split the data set? - Which standards for data and models storage/interchange should I follow? Submitted papers will be reviewed according to the ESANN reviewing process and will be evaluated on their scientific value; originality, correctness, and writing style. 3. Neural Networks for Computational Neuroscience ----------------------------------------------------------------------- Organized by: David Meunier, H?l?ne Paugam-Moisy (LIRIS-CNRS, France) Computational Neuroscience aims at explaining the experimental measurements obtained in electrophysiology (both in animals with intra-cranial recordings and in humans with techniques such as EEG, MEG) by means of models. The models can take several forms, one of them being neural networks. The aim of this session is to contribute to this specific use of neural networks. The way to develop models to explain electrophysiology can have two directions. A first one, qualified of "bottom-up" approach, where the modelling consists in trying to reproduce properties observed at macroscopic level by fitting neuron models (e.g. Hodgkin and Huxley neuron, spiking neuron, integrate-and-fire neuron) and parameters to the measured behaviour of biological neurons. The architecture of the network is also based on experimental anatomical data (e.g. olfactory bulb architecture, hippocampus architecture, etc...). This direction includes mean-field approaches, where the influences of each parameter on the dynamics are studied, and qualitative approaches, aiming at defining the minimal set of properties that are necessary to observe a given dynamical behaviour. A second one is qualified of "top-down" approach, consisting in using evolutionary algorithms to let emerge adapted neural networks with regard to a given task and to study their properties a posteriori. In this case, the network is studied a posteriori to detect why adapted networks are better than initial random networks. The network emergent properties can be studied, by example, with the tools of the theory of complex systems. 4. Machine learning methods in cancer research ----------------------------------------------------------------------- Organized by: Alfredo Vellido (Polytechnic University of Catalonia, Spain), Paulo J.G.Lisboa (Liverpool John Moores University, United Kingdom) Neural Networks and Machine Learning methods in general are widely used in cancer research and published in clinical, as well as methodological journals. Their acceptance among medical practitioners is steadily increasing, in part because of demands for advanced data analysis relating to bioinformatics, but also because of a realization that decision support will be inherent in the current agenda for personalized medicine. The application of Machine Learning to medical data may be said to have entered a period of adolescence, where the early excitement about their potential has been tempered by the need to assure generality through the use of principled approaches to complexity control. The excitement that was communicated during the early phase of development in the late 90?s seems to have whetted the appetite of clinicians for what these methods can achieve, initiating close and fruitful collaborations where key clinical questions are driving new data-based studies, so building clinical relevance, rather than obsolescence, into study design. Machine Learning methods can be applied to a wide range of data types and problems in cancer research. The range of applications includes exploratory analysis and predictive inference, with topics ranging from clustering, through classification, survival analysis, and rule extraction. Hot topics include knowledge discovery from data, but also the integration of multimodal data into clinical inference systems, the use of graphical models for structure finding in large sparse data sets, and methods for robust performance estimation which include the use of automatic rule extraction methods to match inference making with clinical expert knowledge. This special session aims to bring together methodological advances and clinical relevant case studies of Machine Learning approaches to cancer diagnosis and prognosis, and oncology-related bioinformatics. ESANN 2008 participants would benefit from the coming together of a number of internationally renowned experts in the field, who would provide their expert view on a broad palette of state-of-the-art theoretical developments and applications. 5. Machine Learning Approches and Pattern Recognition for Spectral Data ----------------------------------------------------------------------- Organized by: Thomas Villmann (Univ. Leipzig, Germany), Erzs?bet Mer?nyi (Rice University, USA), Udo Seiffert (IPK Gatersleben, Germany) Analysis of spectral data plays an important role in many areas of research like physics, astronomy and geophysics, chemistry, bioinformatics, biochemistry engineering, and others. The amount of data may range from several billion samples in geophysics to only a few in medical applications. Further, a vectorial representation of spectra typically leads to huge-dimensional problems. However, spectral vectors are functional, i.e., the vector dimensions are not independent. The locations, widths and shapes of characteristic peaks or valleys (absorptions), as well as their co-occurences are important for data analyses. These properties should be used for specific machine learning approaches designed for spectral analysis. This special session seeks contributions which report about new developments in this field of research: both outstanding applications using specific techniques and methodologies in machine learning and neural networks for spectral data, as well as new theoretical developments are solicited. The session is intended to cover a broad range of application areas as outlined in the beginning. A possible (non-exhaustive) list of applications/problems could be: - NMR-based applications in physics, chemistry and biology - Remote sensing in astronomy and geophysics - Chemometrics - Bioinformatics - Medical applications - Special techniques for utilization of data-intrinsic dependencies - High-dimensional data and sparseness - Special metrics for data similarities ======================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - Machine Learning Group 3, pl. du Levant - B-1348 Louvain-la-Neuve - Belgium tel: +32 10 47 25 51 - fax: + 32 10 47 25 98 mailto:esann at dice.ucl.ac.be * Conference secretariat d-side conference services 24 av. L. Mommaerts - B-1140 Evere - Belgium tel: + 32 2 730 06 11 - fax: + 32 2 730 06 00 mailto:esann at dice.ucl.ac.be ======================================================== From shimon.whiteson at gmail.com Mon Oct 15 04:59:19 2007 From: shimon.whiteson at gmail.com (Shimon Whiteson) Date: Mon, 15 Oct 2007 10:59:19 +0200 Subject: Connectionists: Second Annual Reinforcement Learning Competition Message-ID: <05D99545-81D6-49C5-95A5-6F29CABFBF02@gmail.com> [Apologies for multiple postings] =============================================== The Second Annual Reinforcement Learning Competition at ICML-08 in Helsinki, Finland Announcement and Call for Participants =============================================== The Second Annual Reinforcement Learning Competition invites researchers from around the world to apply their latest methods to a suite of exciting and diverse challenge problems. The aim of the competition is to facilitate direct comparisons between various learning methods on important and realistic domains. We believe such a competition can stimulate the development and verification of increasingly practical algorithms. This year's event will feature well-known benchmark domains as well as more challenging problems of real-world complexity. The competition domains are: -Mountain Car: Perhaps the most well-known reinforcement learning benchmark task, in which an agent must learn how to drive an underpowered car up a steep mountain road. -Tetris: The hugely popular video game, in which four-block shapes must be manipulated to form complete lines when they fall. -Helicopter Hovering: A simulator, based on the work of Andrew Ng and collaborators, which requires an agent to learn to control a hovering helicopter. -Keepaway: A challenging task, based on the RoboCup soccer simulator, that requires a team of three robots to maintain possession the ball while two other robots attempt to steal it. -Real-Time Strategy: A game, based on popular real-time strategy games, which poses exciting new challenges for the reinforcement learning community. -Polyathlon: The agent will face a set of potentially unrelated MDPs with minimal task knowledge and no prior training. In addition, this year's competition will utilize new evaluation paradigms designed to encourage algorithms that generalize well to previously unseen tasks. In particular, each domain will be paramaterized and competition parameters will differ from those seen by the participants while developing their entries. As a result, only learning algorithms that are robust across a range of parameters can expect to perform well in the competition. The competition will end with an event at the 2008 International Conference on Machine Learning in Helsinki, Finland, at which the winners will be announced. Competitors will be invited to attend and present their methods. The event will also feature invited speakers and discussions about the best way to perform empirical comparisons in reinforcement learning and the future of the competition. Website ====== To learn more about the competition, details of the domains, and how to get started, visit the competition website at: http://rl-competition.org Timeline ======= 1 November, 2007: Public release of competition training software 1 December, 2007: Public release of competition test software; competitors can begin submitting results 1 July 2008: Deadline to submit competition results 6-9 July 2008: Competition event at ICML in Helsinki, Finland Organizing Committee ================== Shimon Whiteson, Universiteit van Amsterdam (Chair) Adam White, University of Alberta Rich Sutton, University of Alberta Doina Precup, McGill University Peter Stone, University of Texas at Austin Michael Littman, Rutgers University Nikos Vlassis, Technical University of Crete Martin Riedmiller, Universit?t Osnabr?ck Technical Committee ================= Brian Tanner, University of Alberta (Chair) Marc Lanctot, University of Alberta Pieter Abbeel, Stanford University Matt Taylor, University of Texas at Austin Shivaram Kalyanakrishnan, University of Texas at Austin Leah Hackman, University of Alberta Mark Lee, University of Alberta Jordan Frank, McGill University Nick Jong, University of Texas at Austin From kenji at ieee.org Mon Oct 15 12:10:44 2007 From: kenji at ieee.org (Kenji Suzuki) Date: Tue, 16 Oct 2007 01:10:44 +0900 Subject: Connectionists: Postdoctoral Positions in Cybernics at the University of Tsukuba Message-ID: <006601c80f45$f14d1560$d3e74020$@org> * We apologize if you receive multiple copies of this announcement. Job announcement Postdoctoral Positions in Cybernics (up to five positions) Cybernics Program, University of Tsukuba Tsukuba, Japan Ref: CYB02/P0710 Applications are invited for five (5) postdoctoral positions in the area of Cybernics: fusion of human, machine and information systems, or relevant subjects. Cybernics is a new domain of interdisciplinary academic field of human-assistive technology to enhance, strengthen, and support human's cognitive and physical functions, which challenges to integrate and harmonize humans and robots (RT: robotics technology) with the basis of information technology (IT). The three primary research areas are: (i) Cybernoid: robot suits, cybernic limb and hand, implanted cybernic system, subjective cognition computing, virtual human-body kernel. (ii) Next-generation interface: brain-computer interface, somato-sensory media, humanoid, medical interface, ubiquitous sensing interface, intelligent robots. (iii) Management technology for next-generation advanced systems: network security, new-generation risk management, cognitive engineering, ethical, sociological, and conceptual readiness. The position is initially for duration of three years, with possible extension to a maximum of four years depending on achieved results. To change the appointment as an acting faculty (associate/assistant professor) may be considered, within the fixed term of three years, upon request. Our primary focus is conducting research and pursuing education at the Ph.D. level in the area of Cybernics. Responsibility may include some administrative tasks related to the program. These positions will be available immediately. The review of applications will proceed after October 29th 2007 on their arrival and will continue until the position is filled. The Cybernics program (Leader, Prof. Yoshiyuki Sankai) is supported by the grant-in-aid science research under the Global COE program, and also is a part of a new strategic initiative in the University of Tsukuba. For more details on research areas, see the program website, http://www.cybernics.tsukuba.ac.jp/. Candidates are expected to have earned a Ph.D. degree or equivalent in a relevant subject area and to have demonstrated achievement in their fields. Candidates recently graduated Ph.D., or expected to obtain their Ph.D. degrees or equivalent before the appointment begins are also encouraged to apply. Applications should include a full curriculum vitae, a list of publications, 5 reprints or preprints of relevant papers, an outline of past researches, future research plans, and possible contribution to this project (about A4 2-pages), and the names and addresses of at least two referees. Applicants are requested, if possible, to list publications under the following main headings: Authored or Edited books, Refereed Journal papers, Review articles, Refereed conference proceedings, Non-refereed conference proceedings, Patents, Awards, External funding and grants, Others. Please return your application by email, preferably in PDF format, to jobs.pd at cybernics.tsukuba.ac.jp. You should include the ref. number you are applying for in the header of your e-mail, or standard mail envelope. For informal inquiries please contact the Leader of the Cybernics Program, Professor Yoshiyuki Sankai on email sankai at cybernics.tsukuba.ac.jp. Tsukuba is a university and science city, located about 60 kilometers, about 1 hour by car, northeast of central Tokyo. Over 50 national and independently administered research organizations are concentrated in the Tsukuba Science City district, which is centered on the university. Review of applications: from 29 October, 2007 * Program website http://www.cybernics.tsukuba.ac.jp/ --- Dr Kenji Suzuki kenji at ieee.org Assistant Professor University of Tsukuba, Japan http://www.iit.tsukuba.ac.jp/~kenji/ From announce at ccnconference.org Fri Oct 19 16:24:36 2007 From: announce at ccnconference.org (announce@ccnconference.org) Date: Fri, 19 Oct 2007 14:24:36 -0600 Subject: Connectionists: Registration DEADLINE extended for CCNC 2007 one week to Fri 26-Oct-07 Message-ID: <200710191424.36750.announce@ccnconference.org> ~ CCNC 2007/DYNAMICAL NEUROSCIENCE XV ~ The planning committee has extended the deadline for standard registration to MIDNIGHT next FRIDAY October 26, 2007. After that, registration fees will increase to $225 faculty/$110 student. The committee requests that you PLEASE try to register for the conference by this deadline in order to help us optimally plan the conference and manage our budget. Registration should be done via the conference website at: http://ccnconference.org/page5.html HOTEL reservations can be made via the SfN website: http://www.sfn.org. (CCNC will be held in the San Diego Convention Center.) The final program schedule for this year's conference can be viewed at: http://ccnconference.org/page2.html PLEASE NOTE: In an effort to be family friendly (Halloween), the first day will start at 1:30PM to allow for travel that morning, if desired. Based on the number of poster submissions, it looks like this year's conference should be very well attended, with lots of very interesting work! Regards, CCNC 2007 Planning Committee ---------------------------------------------------------------------------- 3RD ANNUAL CONFERENCE ON COMPUTATIONAL COGNITIVE NEUROSCIENCE ????????????????www.ccnconference.org To be held in conjunction with Dynamical Neuroscience XV immediately prior to the 2007 SOCIETY FOR NEUROSCIENCE (SfN) meeting, November 3-7, 2007 at the San Diego Convention Center, San Diego, CA. * CONFERENCE DATES: Thu-Fri November 1 & 2, 2007 * REGISTER AT: ????????www.ccnconference.org/page5.html The inaugural CCNC 2005 meeting held prior to Society for Neuroscience (SfN) in Washington, DC (also in conjunction with the Dynamical Neuroscience satellite) was a great success, with approximately 250 attendees, 60 presented posters, and strongly positive reviews. For 2006, we went to Houston for the much smaller Psychonomics meeting and still had over 100 attendees and almost 50 posters. In future years, we will continue to rotate among different neuroscience and psychology meetings. ____________________________________________________________________________ * DEADLINE FOR SUBMISSION OF ABSTRACTS: (NOW CLOSED - POSTERS ACCEPTED ON A SPACE AVAILABLE BASIS ONLY) Abstracts are to be submitted online via the website: ???????? ?www.ccnconference.org/page6.html As in past years, there will be two categories of submissions: ? ?-Poster only ? ?-Poster, plus short talk (15 min) to highlight the poster Abstracts should be limited to 250 words. Women and underrepresented minorities are especially encouraged to apply. Reviewing of posters will be inclusive and only to ensure appropriateness to the meeting. Short talks will be selected on the basis of research quality, relevance to conference theme, and expected accessibility in a talk format. Abstracts not selected for short talks will still be accepted as posters as long as they meet appropriateness criteria. * NOTIFICATION OF POSTER ACCEPTANCE: DONE * CONTRIBUTED SHORT TALK SELECTION: DONE __________________________________________________________________________ Program Summary: * 2007 Keynote Speakers: ? ? ? ? Read Montague, Baylor College of Medicine Alex Pouget, University of Rochester ? ? ? * 3 Symposia, each including a mixture of modelers and non-modelers and ? focused on a common theme or issue: ** Computational Models in Biological Psychiatry Moderator: Michael Frank, University of Arizona ** Computionally-Based Brain Imaging: Models, Levels, and Approaches ? ?Moderator: Todd Braver, Washington University - St. Louis ** Hippocampal Neurogenesis in Learning and Memory ? ?Moderator: Janet Wiles, University of Queensland * 12 short talks will be chosen featuring selected posters * Poster sessions ____________________________________________________________________________ 2007 Planning Committee: Suzanna Becker, McMaster University Jonathan Cohen, Princeton University Nathaniel Daw, New York University David Noelle, University of California, Merced Maximilian Riesenhuber, Georgetown University Medical Center Randall O'Reilly, University of Colorado, Boulder (ex officio) Executive Staff: Thomas Hazy, University of Colorado, Boulder For more information and to sign up for the mailing list visit: ???????? www.ccnconference.org _______________________________________________ From joaquinc at microsoft.com Wed Oct 10 04:21:35 2007 From: joaquinc at microsoft.com (=?iso-8859-1?Q?Joaquin_Qui=F1onero_Candela?=) Date: Wed, 10 Oct 2007 09:21:35 +0100 Subject: Connectionists: NIPS*07 Workshop: Machine Learning and Games (MALAGA) Message-ID: Announcement and call for contributions: ===================================================== NIPS 2007 Workshop on Machine Learning and Games (MALAGA) http://research.microsoft.com/mlp/apg/malaga.aspx Saturday 8 of December 2007, Whistler BC ===================================================== ---- Organizers: Joaquin Qui?onero Candela Thore Graepel Ralf Herbrich Applied Games, Microsoft Research Cambridge ---- ---- Description: Computer games sales are three time larger than industry software sales, and on par with Hollywood box office sales; Halo 3 has become the biggest entertainment launch in history, with $170 Million in sales in the US alone in the first 24 hours! Modern computer games are often based on extremely complex simulations of the real world and constitute one of the very few real fields of application for artificial intelligence encountered in everyday life. Surprisingly, machine learning methods are not present in the vast majority of computer games. There have been a few recent and notable successes in turn-based two-player, discrete action space games such as Backgammon, Checkers, Chess and Poker. However, these successes are in stark contrast to the difficulties still encountered in the majority of computer games, which typically involve more than two agents choosing from a continuum of actions in complex artificial environments. Typical game AI is still largely built around fixed systems of rules that often result in implausible or predictable behaviour and poor user experience. The purpose of this workshop is to involve the NIPS community in the exciting challenges that games - ranging from traditional table top games to cutting-edge console and PC games - offer to machine learning. ---- ---- Call for Contributions: We invite the submission of relevant work for oral presentation at the workshop. Format: 2 page long extended abstract in either Letter or A4. Deadline: Friday November 2nd, 2007. Talk duration: 15 minutes for the presentation, 5 minutes for discussion. ---- ------------- Joaquin Qui?onero Candela, Associate Researcher, Applied Games Group, Microsoft Research Ltd., 7 J J Thomson Avenue, Cambridge, CB3 0FB, UK, Tel: +44 (0)1223 479 700, Fax: +44 (0)1223 479 999 From terry at salk.edu Sat Oct 20 04:38:31 2007 From: terry at salk.edu (Terry Sejnowski) Date: Sat, 20 Oct 2007 01:38:31 -0700 Subject: Connectionists: Neuro Thursday at NIPS In-Reply-To: Message-ID: NEURO-THURSDAY at NIPS: Thusday, December 6 - 8:30 AM - Noon - Vancouver Thursday, December 6 (the final day of the Conference), will be devoted to Neuroscience, and will consist of a fascinating invited talk by Professor Manabu Tanifuji (Riken) on the monkey visual cortex, plus six outstanding plenary talks. In addition, all of the Neuroscience posters will take place on Wednesday night, allowing early arrivals to interact with researchers. The Wednesday night poster program will also contain many posters on Machine learning and Computer Vision, focused on topics that are also relevant to Neuroscience. All of the morning events (including the Wednesday night Poster Session and the Spotlights that precede it) will be available for the special "Neuro-Thursday" registration rate of $50. For those attending the entire Conference, "Neuro-Thursday" is included in the registration price. -------- Deep Learning Workshop: Foundations and Future Directions Thursday, December 6 - 2:00 to 5:30 PM - Vancouver Theoretical results strongly suggest that in order to learn the kind of complicated functions that can represent high-level abstractions (e.g. in vision, language, and other AI-level tasks), one may need "deep architectures", which are composed of multiple levels of non-linear operations (such as in neural nets with many hidden layers). Searching the parameter space of deep architectures is a difficult optimization task, but learning algorithms (e.g. Deep Belief Networks) have recently been proposed to tackle this problem with notable success, beating the state-of-the-art in certain areas. This Workshop is intended to bring together researchers interested in the question of deep learning in order to review the current algorithms' principles and successes, but also to identify the challenges, and to formulate promising directions of investigation. Besides the algorithms themselves, there are many fundamental questions that need to be addressed: What would be a good formalization of deep learning? What new ideas could be exploited to make further inroads to that difficult optimization problem? What makes a good high-level representation or abstraction? What type of problem is deep learning appropriate for? There is no charge for this Workshop or for the bus to Whistler that will leave after the Workshop; however, a separate registration is required. To register: http://www.iro.umontreal.ca/~lisa/twiki/bin/view.cgi/Public/DeepLearningWorkshopNIPS2007 --------- NIPS Whistler Workshops Friday December 7 - Saturday December 8, 2007 The post-Conference Workshops will be held at the Westin Resort and Spa and the Westin Hilton in Whistler, British Columbia, Canada on December 7 and 8, 2007. The Workshops provide multi-track intensive sessions on a wide range of topics. The venue and schedule facilitate informality and depth. Partial List of Workshop Topics and Organizers: Beyond Simple Cells: Probabilistic Models for Visual Cortical Processing Richard Turner, Pietro Berkes, Maneesh Sahani Hierarchical Organization of Behavior: Computational, Psychological and Neural Perspectives Yael Niv, Matthew Botvinick, Andrew Barto Large Scale Brain Dynamics Ryan Canolty, Kai Miller, Joaquin Quionero Candela, Thore Graepel, Ralf Herbrich Mechanisms of Visual Attention Jillian Fecteau, Dirk Walther, Vidhya Navalpakkam, John Tsotsos Music, Brain and Cognition. Part 1: Learning the Structure of Music and Its Effects On the Brain David Hardoon, Eduardo Reck-Miranda, John Shawe-Taylor Music, Brain and Cognition. Part 2: Models of Sound and Cognition Hendrik Purwins, Xavier Serra, Klaus Obermayer Principles of Learning Problem Design John Langford, Alina Beygelzimer Representations and Inference on Probability Distributions Kenji Fukumizu, Arthur Gretton, Alex Smola The Grammar of Vision: Probabilistic Grammar-Based Models for Visual Scene Understanding and Object Categorization Jan Peters, Marc Toussaint A complete list of all 25 workshop and links for more information: http://nips.cc/Conferences/2007/Program/schedule.php?Session=Workshops --------- From mail at jan-peters.net Mon Oct 22 02:54:49 2007 From: mail at jan-peters.net (Jan Peters) Date: Mon, 22 Oct 2007 06:54:49 -0000 Subject: Connectionists: [NIPS 2007 WORKSHOP] Abstract Submission Extended! Reminder/Call for Posters-Robotics Challenges for Machine Learning Message-ID: <6804218E-EC2B-4F45-83A6-38A1E6D15E4C@jan-peters.net> *** Apologies for Multiple Postings *** Abstract Submission Extended! Abstract Submission Extended! Abstract Submission Extended! Abstract Submission Extended! Abstract Submission Extended! Abstract Submission Extended! ======== ==== CALL FOR POSTERS ==== =========== NIPS 2007 WORKSHOP: Robotics Challenges for Machine Learning Dates: 7 December, 2007 Organizers: Jan Peters (Max Planck Institute for Biological Cybernetics & USC), Marc Toussaint (Technical University of Berlin) WWW: http://www.robot-learning.de email: nips07 at robot-learning.de Poster request: Submit a one page abstract!!! Abstract Submission Deadline: October 24, 2007 Acceptance Notification: October 26, 2007 ======== ==== CALL FOR POSTERS ==== =========== Abstract: Creating autonomous robots that can assist humans in situations of daily life is a great challenge for machine learning. While this aim has been a long standing vision of robotics, artificial intelligence, and the cognitive sciences, we have yet to achieve the first step of creating robots that can accomplish a multitude of different tasks, triggered by environmental context or higher level instruction. Despite the wide range of machine learning problems encountered in robotics, the main bottleneck towards this goal has been a lack of interaction between the core robotics and the machine learning communities. To date, many roboticists still discard machine learning approaches as generally inapplicable or inferior to classical, hand-crafted solutions. Similarly, machine learning researchers do not yet acknowledge that robotics can play the same role for machine learning which for instance physics had for mathematics: as a major application as well as a driving force for new ideas, algorithms and approaches. Some fundamental problems we encounter in robotics that equally inspire current research directions in Machine Learning are: -- learning and handling models, (e.g., of robots, task or environments) -- learning deep hierarchies or levels of representations (e.g., from sensor & motor representations to task abstractions) -- regression in very high-dimensional spaces for model and policy learning -- finding low-dimensional embeddings of movement as an implicit generative model -- methods for probabilistic inference of task parameters from vision, e.g., 3D geometry of manipulated objects -- the integration of multi-modal information (e.g., proprioceptive, tactile, vision) for state estimation and causal inference -- probabilistic inference in non-linear, non-Gaussian stochastic systems (e.g., for planning as well as optimal or adaptive control) Robotics challenges can inspire and motivate new Machine Learning research as well as being an interesting field of application of standard ML techniques. Inversely, with the current rise of real, physical humanoid robots in robotics research labs around the globe, the need for machine learning in robotics has grown significantly. Only if machine learning can succeed at making robots fully adaptive, it is likely that we will be able to take real robots out of the research labs into real, human inhabited environments. To do so, future robots will need to be able to make proper use of perceptual stimuli such as vision, proprioceptive & tactile feedback and translate these into motor commands. To close this complex loop, machine learning will be needed on various stages ranging from sensory-based action determination over high-level plan generation to motor control on torque level. Among the important problems hidden in these steps are problems which can be understood from the robotics and the machine learning point of view including perceptuo-action coupling, imitation learning, movement decomposition, probabilistic planning problems, motor primitive learning, reinforcement learning, model learning and motor control. Format: The goal of this one-day workshop is to bring together people that are interested in robotics as a source and inspiration for new Machine Learning challenges, or which work on Machine Learning methods as a new approach to robotics challenges. In the robotics context, among the questions which we intend to tackle are Reinforcement Learning, Imitation, and Active Learning: * What methods from reinforcement learning scale into the domain of robotics? * How can we improve our policies acquired through imitation by trial and error? * Can we turn many simple learned demonstrations into proper policies? * Does the knowledge of the cost function of the teacher help the student? * Can statistical methods help for generating actions which actively influencing our perception? E.g., Can these be used to plan visuo-motor sequences that will minimize our uncertainty about the scene? * How can image understanding methods be extended to provide probabilistic scene descriptions suitable for motor planning? Motor Representations and Control: * Can we decompose human demonstrations into elemental movements, e.g., motor primitives, and learn these efficiently? * Is it possible to build libraries of basic movements from demonstration? How to create higher-level structured representations and abstractions based on elemental movements? * Can structured (e.g., hierarchical) temporal stochastic models be used to plan the sequencing and superposition of movement primitives? * Is probabilistic inference the road towards composing complex action sequences from simple demonstrations? Are superpositions of motor primitives and the coupling in timing between these learnable? * How to generate compliant controls for executing complex movement plans which include both superposition and hierarchies of elemental movements? Can we find learned versions of prioritized hierarchical control? * Can we learn how to control in task-space of redundant robots in the presence of under-actuation and complex constraints? Can we learn force or hybrid control in task-space? * Is real-time model learning the way to cope with executing tasks on robots with unmodeled nonlinearities and manipulating uncertain objects in unpredictable environmental interactions? * What new regression techniques can help real-time model learning to improve the execution of tasks on robots with unmodeled nonlinearities and manipulating uncertain objects in unpredictable environmental interactions? Learning structured models and representations: * What kind of probabilistic models provide a compact and suitable description of real-world environments composed of manipulable objects? * How can abstractions or compact representations be learnt from sensori-motor data? * How can we extract features of the sensori-motor data that are relevant for motor control or decision making? E.g., can we extract visual features of objects directly related to their manipulability or ``affordance''? Posters: We are open for any posters posing problems for machine learning and for presenting machine learning algorithms with applications in robotics. Please send us a one page abstract (A4 or letter) describing the poster which you intend to present with at least one reference to previous work. Choose a format of your liking, e.g., the standard NIPS template. The deadline for abstract submissions is October 24, 2007 and the notification will be October 26, 2007 Abstract Submission Deadline: October 24, 2007 Acceptance Notification: October 26, 2007 From terry at salk.edu Wed Oct 24 22:37:48 2007 From: terry at salk.edu (Terry Sejnowski) Date: Wed, 24 Oct 2007 19:37:48 -0700 Subject: Connectionists: NEURAL COMPUTATION - November 2007 In-Reply-To: Message-ID: Neural Computation - Contents - Volume 19, Number 11 - November 1, 2007 Notes: J4 at Sweet 16: A New Wrinkle? Bardia Behabadi and Bartlett Mel On the Consistency of Bayesian Function Approximation Using Step Functions Heng Lian Letters: Learning Real World Stimuli in a Neural Network with Spike-Driven Synaptic Dynamics Stefano Fusi, Joseph M. Brader, and Walter Senn Tight Data-Robust Bounds to Mutual Information Combining Shuffling and Model Selection Techniques Stefano Panzeri, Marcelo A. Montemurro, and Riccardo Senatore Spike-Frequency Adapting Neural Ensembles: Beyond Mean-Adaptation and Renewal Theories Eilif Muller, Lars Buesing, Johannes Schemmel, and Karlheinz Meier Transition and Hysteresis in an Ensemble of Stochastic Spiking Neurons Andreas Kaltenbrunner, Vicenc Gomez , and Vicente Lopez Model-Based Reinforcement Learning for Partially Observable Games with Sampling-Based State Estimation Shin Ishii and Hajime Fujita Clustering Based on Gaussian Processes Jaewook Lee and Hyun-Chul Kim Elman Backpropagation as Reinforcement for Simple Recurrent Networks Andre Gruning ON-LINE - http://neco.mitpress.org/ SUBSCRIPTIONS - 2007 - VOLUME 19 - 12 ISSUES Electronic only USA Canada* Others USA Canada* Student/Retired $60 $63.60 $114 $54 $57.24 Individual $100 $106.00 $154 $90 $95.40 Institution $782 $828.92 $836 $704 $746.24 * includes 6% GST 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 bengioy at iro.umontreal.ca Thu Oct 25 17:32:30 2007 From: bengioy at iro.umontreal.ca (Yoshua Bengio) Date: Thu, 25 Oct 2007 17:32:30 -0400 Subject: Connectionists: NIPS satellite meeting on deep learning References: Message-ID: <8FF89F80-82D1-45A3-81ED-C3E2935C6241@iro.umontreal.ca> Hello, Following up on Geoffrey Hinton's Monday NIPS tutorial on Deep Belief Nets, there will be a satellite meeting/workshop on Thursday afternoon (i.e. not at the usual time for workshops) at the Vancouver Hyatt. This Deep Learning workshop http://www.iro.umontreal.ca/~lisa/deepNIPS2007 is part of the "Neuro-Thursday" this year http://nips.cc/Conferences/2007/Neuro-Thursday SEATING IS LIMITED at the Hyatt for this workshop, so if you are interested (check the program at the URL above) you are encouraged to register as soon as possible. For interested participants of the deep learning workshop buses to Whistler are scheduled after the workshop (the regular buses to Whisler are at 2pm). The workshop also honors Geoffrey Hinton's 60-th birthday. More on the meeting: Theoretical results strongly suggest that in order to learn the kind of complicated functions that can represent high-level abstractions (e.g. in vision, language, and other AI-level tasks), one may need "deep architectures", which are composed of multiple levels of non- linear operations (such as in neural nets with many hidden layers). Searching the parameter space of deep architectures is a difficult optimization task, but learning algorithms (e.g. Deep Belief Networks) have recently been proposed to tackle this problem with notable success, beating the state-of-the-art in certain areas. This workshop is intended to bring together researchers interested in the question of deep learning in order to review the current algorithms' principles and successes, but also to identify the challenges, and to formulate promising directions of investigation. Besides the algorithms themselves, there are many fundamental questions that need to be addressed: What would be a good formalization of deep learning? What new ideas could be exploited to make further inroads to that difficult optimization problem? What makes a good high-level representation or abstraction? What type of problem is deep learning appropriate for? Schedule: 2:00pm - 2:25pm Yee-Whye Teh, Gatsby Unit : Deep Belief Networks 2:25pm - 2:45pm John Langford, Yahoo Research: Theoretical Results on Deep Architectures 2:45pm - 3:05pm Yoshua Bengio, University of Montreal: Optimizing Deep Architectures 3:05pm - 3:25pm Yann Le Cun, New York University: Learning deep hierarchies of invariant features 3:25pm - 3:45pm Martin Szummer, Microsoft Research: Deep networks for information retrieval 3:45pm - 4:00pm Coffee break 4:00pm - 4:20pm Max Welling, University of California: Hierarchical Representations from networks of HDPs 4:20pm - 4:40pm Andrew Ng, Stanford University: Self-taught learning: Transfer learning from unlabeled data 4:40pm - 5:10pm Geoff Hinton, University of Toronto: Restricted Boltzmann machines with multiplicative interactions 5:10pm - 5:30pm Discussion -- Yoshua Bengio From odobez at idiap.ch Tue Oct 23 05:51:49 2007 From: odobez at idiap.ch (Jean-Marc Odobez) Date: Tue, 23 Oct 2007 11:51:49 +0200 Subject: Connectionists: Postdoc position in vision - IDIAP Research Institute, CH Message-ID: <471DC435.8040305@idiap.ch> The IDIAP Research Institute, an EPFL laboratory (Swiss Federal Institute of Technology, Lausanne) seeks immediately one qualified postdoctoral researcher in the field of computer vision and machine learning. The postdoc will work in the framework of the AMI/AMIDA EU project jointly lead by IDIAP and the University of Edinburgh, which addresses the general issue of computer enhanced multi-modal interaction in the context of meetings. The postdoc research will address the issue of Non-Verbal communication analysis in meetings, with a focus on the design of a close to real-time Head tracking and Pose estimation algorithm to be used for estimating the Visual Focus of Attention of meeting participants. The research will build on several years of experience gained by IDIAP on this topic, and will be conducted in collaboration with researchers working on Human-Centered computing. The successful candidate will have a PhD and experience in at least 2 of the following fields: - computer vision or image processing - machine learning, pattern recognition or statistical techniques - tracking, preferably within a probabilistic framework. In addition, the candidate should have a strong background in C++ programming, preferably under a Linux environment, and be interested in working towards a live demonstrator of the investigated techniques and models. Experience in face modeling would be an asset. The applicant should have good demonstrated skills in written and spoken English. The initial Postdoctoral position is for one year, with a possibility of 1 year extension. Annual gross salary ranges from ChF 65000 to ChF 75000 (depending on qualifications and experience). The position is open immediately. Interested candidates should send a letter of motivation, along with their detailed CV, electronic transcripts of B.S. and M.S. degrees and names of 3 references to jobs at idiap.ch. Please precise the adequate job reference (AMIDA-VFOA). More information can also be obtained by contacting Jean-Marc Odobez (odobez at idiap.ch), or Alex Jaimes (ajaimes at idiap.ch). ----------------------------- References: - website IDIAP: http://idiap.epfl.ch or http://www.idiap.ch - website AMI/AMIDA: http://www.amiproject.org/ami-scientific-portal ----------------------------- About IDIAP: IDIAP is an equal opportunity employer and is actively involved in the European initiative involving the Advancement of Women in Science. IDIAP seeks to maintain a principle of open competition (on the basis of merit) to appoint the best candidate, provides equal opportunity for all candidates, and equally encourages both females and males to consider employment with IDIAP. Although IDIAP is located in the French part of Switzerland, English is the main working language. Free French lessons are provided. IDIAP is located in the town of Martigny in Valais, a scenic region in the south of Switzerland, surrounded by the highest mountains of Europe, and offering exciting recreational activities, including hiking, climbing and skiing, as well as varied cultural activities. It is within close proximity to Montreux (Jazz Festival) and Lausanne. From py at csl.sony.fr Thu Oct 25 06:38:42 2007 From: py at csl.sony.fr (Pierre-Yves Oudeyer) Date: Thu, 25 Oct 2007 12:38:42 +0200 Subject: Connectionists: [urgent] INRIA postdoc proposal in developmental robotics Message-ID: <47207232.8090304@csl.sony.fr> Dear colleagues, as part of the creation of a developmental/epigenetic robotics group in INRIA Bordeaux, an open postdoc position has just become available. This position would last one year, start in december or january, be located in Bordeaux and supervised by myself. There are four possible topics: 1) development of advanced curiosity-driven learning algorithms and experiments on robots, 2) development of techniques for achieving naturally and intuitively joint attention between a robot and a human, 3) modeling of developmental mechanisms for language acquisition in robots and human, 4) user/human factor studies in entertainment-robot teaching/taming interactions. The constraint is that candidatures must be done ASAP. Candidates are expected to have already produced international level publications in relevant journals and conferences during their PhD, and are also expected to have had a significant experience of practical computational experiments in robotics, machine learning or interaction design (except for the user studies prosposal). Potential interested candidates can send me an email for further information, Best regards, Pierre-Yves Oudeyer http://www.csl.sony.fr/~py email: py at csl.sony.fr From erik at oist.jp Fri Oct 26 01:59:14 2007 From: erik at oist.jp (Erik De Schutter) Date: Fri, 26 Oct 2007 14:59:14 +0900 Subject: Connectionists: High performance biological computing workshop at OIST Message-ID: The Okinawa Institute of Science and Technology organizes a workshop on "Hardware and software for large-scale biological computing in the next decade", December 11-14, 2007, Okinawa, Japan. Speakers: Phil ANDREWS Alan GARA Robert GROSSMAN Seth Copen GOLDSTEIN Mike HINES Tetsuya SATO Felix SCHUERMANN Masakazu SEKIJIMA John SHALF Thomas STERLING Arthur TREW Tadashi WATANABE John WAWRZYNEK Full program and details can be found at http://www.irp.oist.jp/hpc- workshop/ The workshop can accept a limited number of additional non-OIST participants. These participants will not be supported by OIST: they will have to cover all costs (travel and accommodation) themselves. Free registration is required and acceptance will be confirmed by email on a first-come, first-served basis. A registration form is available on the website. From erik at oist.jp Fri Oct 26 02:10:50 2007 From: erik at oist.jp (Erik De Schutter) Date: Fri, 26 Oct 2007 15:10:50 +0900 Subject: Connectionists: Positions in reaction-diffusion modeling of signaling pathways involved in synaptic plasticy Message-ID: <6BBFDFF0-556E-459F-BB60-A4EA61B30F29@oist.jp> Postdoctoral position in modeling of signaling pathways involved in synaptic plasticy A postdoctoral position is available in the Computational Neuroscience Unit of Dr. Erik De Schutter at the Okinawa Institute of Science and Technology (http://www.irp.oist.jp/cns/) to study the signaling pathways involved in cerebellar long-term depression at the parallel fiber to Purkinje cell synapse. Main goal is to apply reaction-diffusion modeling to better understand how these signaling pathways operate under stochastic conditions in the presence of strong concentration gradients. Additional topics are inclusion of other forms of synaptic plasticity and of downstream signaling pathways in the models and to study the effect of anomalous diffusion on these pathways (Santamaria et al., Neuron 2:635-48, 2006). Candidates should have experience with modeling signaling pathways. There will be ample opportunity to interact with other modelers in the lab who are working on several Purkinje cell and cerebellar related projects and with other scientists at OIST. We offer attractive financial and working conditions in an English language environment. Duration of initial appointments will depend on previous experience, appointments can last up to 4 years and start begin 2008. More information about the Okinawa Institute of Science and Technology is available at http://www.oist.jp/ Send curriculum vitae, summary of research interests and experience, and the names of three referees to Dr. Erik De Schutter at erik at oist.jp --------------- Scientific programming/PhD position in stochastic reaction-diffusion modeling A scientific programming/PhD position position is available in the Computational Neuroscience Unit of Dr. Erik De Schutter at the Okinawa Institute of Science and Technology (http://www.irp.oist.jp/ cns/) to continue programming of the STEPS software. STEPS implements stochastic reaction-diffusion modeling as C++ implementation of the Gillespie Direct Method in a tetrahedral mesh and is controlled with a Python interface (Santamaria et al., Neuron 2:635-48, 2006). Future work involves parallelization, introduction of approximate methods (tau-leaping, Langevin equation), improvement of mesh generation, etc. The candidate should have extensive programming experience, preferentially a degree in computer science, and have sufficient scientific training to understand the mathematical basis of the modeling methods used. We require an active interest in the scientific domain so that the software design can be further developed autonomously. The position can be filled either as a technician or to obtain a PhD degree in biomedical sciences. There will be ample opportunity to interact with modelers in the lab who use the STEPS other modeling software. We offer attractive financial and working conditions in an English language environment. Duration of initial appointments will depend on previous experience, appointments can last up to 4 years and start begin 2008. More information about the Okinawa Institute of Science and Technology is available at http://www.oist.jp/ Send curriculum vitae, summary of research interests and experience, and the names of three referees to Dr. Erik De Schutter at erik at oist.jp From dnoelle at ucmerced.edu Sat Oct 27 19:13:39 2007 From: dnoelle at ucmerced.edu (David C. Noelle) Date: Sat, 27 Oct 2007 16:13:39 -0700 Subject: Connectionists: Assistant Professor of Cognitive Engineering Message-ID: <1193526820.4945.22.camel@69-44-244-2.mrc.clearwire-dns.net> ASSISTANT PROFESSOR OF COGNITIVE ENGINEERING UNIVERSITY OF CALIFORNIA, MERCED The University of California at Merced is accepting applications for Assistant Professor of Cognitive Engineering, as part of a long-term plan to build a concentration of interdisciplinary researchers at the intersection of cognitive science and computer science. Successful applicants will have expertise in both cognitive science and computer science, and will have primary research interests in at least one of the following domains: human-computer interaction, human-robot interaction, user interfaces, augmented or virtual reality, computer games, computational approaches to group cognition, and/or distributed intelligent systems. A Ph.D. in computer science, cognitive science, or a closely related field, is required. Applicants are expected to have strong publication records, clear interdisciplinary research programs, and the ability to attract extramural funding. This position will involve a joint appointment between the School of Engineering and the School of Social Sciences, Humanities, and Arts. Further information may be found at "www.ucmerced.edu/jobs/". Applications must be received by January 15, 2008. From jean-marc.bollon at inrialpes.fr Mon Oct 22 10:08:11 2007 From: jean-marc.bollon at inrialpes.fr (Jean-Marc Bollon) Date: Mon, 22 Oct 2007 16:08:11 +0200 Subject: Connectionists: BAYESIAN COGNITION Winter School Message-ID: ----------------------------------------------------- BAYESIAN COGNITION Winter School ----------------------------------------------------- January 6-11, 2008 Chamonix - Mont-Blanc, France Web : http://bayesian-cognition.org Rationale: -------------- Animals and artificial systems alike are faced with the problem of making inferences about their environments and choosing appropriate responses based on incomplete, uncertain and noisy data. Probabilistic models and algorithms are flourishing in both life sciences and information sciences as ways of understanding the behavior of subjects and the neural processing underlying this behavior, and building robots and artificial agents that can function effectively in such circumstances. The objective of this winter school is to present the latest advances in this subject. This winter school is a prolongation of the Bayesian Cognition workshop held in Paris in January 2006 (Bayesian-Cognition.org) Program: ------------- The school is made of 7 tutorial modules (4 hours long) covering the following topics made by senior researchers in the field: - Probability theory as an alternative to logic (Pierre Bessi?re - CNRS, Grenoble) - Satistical learning (Samy Bengio - Google, Mountain View) - Probabilistic models of Central Nervous System (Sophie Den?ve - Ecole Normal Sup?rieur, Paris) - Approximate evaluation of Bayesian calculus (Vaclav Smidl - UTIA, Prague) - Probabilistic Robotics (Wolfram Burgard - Universit?t Freiburg) - Probabilistic interpretation of physiological and psychophysical data (Jacques Droulez - Coll?ge de France, Paris) - Industrial applications (Emmanuel Mazer - ProBAYES, Grenoble) There will be also 9 presentations (50 minutes long) made by young researchers (typically Postdocs) who will present their thesis work in some details. Finally, there will be a 6 hours practical training session where students will have the opportunity to design, write and run short Bayesian programs using the ProBT? toolkit (Juan-Manuel Ahuactzin - ProBAYES, Grenoble) Main clientele: -------------------- The winter school is mainly conceived for PhD students in cognitive science and robotics interested by the variety of probabilistic models and techniques used in these fields. Master students and postdocs could also find benefits in attending this school. Accomodation: -------------------- The accommodation will be at the Centre Jean Franco in Chamonix, France: http://www.centrejeanfranco.asso.fr/ The Centre Jean Franco is the old national school of mountaineering. The school moved to new buildings in Chamonix. It is located downtown Chamonix, 200 hundred meters from the Aiguille du midi cable car. It offer accommodation for at least 60 students by room of 3. There is a big amphitheatre for the venue of the tutorial. Evaluation and admission criteria: ---------------------------------------------- PhD students and PostDocs: Admission will be done on a first registered basis for PhD and PostDocs students in cognitive science and robotics. Master students: will be accepted only if some spare room is still available on December 1st 2007. PostDoc Call for presentation: ----------------------------------------- There will be 9 presentations (50 minutes long) by PostDocs or young researchers in the field who will have the opportunity to present in details their PhD work. Please send a short abstract of your work before November 30th, 2007. The selected persons will have no registration fee to pay. Registration: ----------------- For registration please go to the Bayesian-Cognition.org web site The registration fee (including lodging and meals for 6 days) is : 500? The winter school is sponsored by the EURON (European Robotics Research Network - http://www.euron.org/) and consequently there is a 200? discount for any student coming from a EURON member institution. Contact Person: ---------------------- Jean-Marc.Bollon at inrialpes.fr From franco at dii.unisi.it Mon Oct 29 11:35:42 2007 From: franco at dii.unisi.it (Franco Scarselli) Date: Mon, 29 Oct 2007 16:35:42 +0100 Subject: Connectionists: Deadline extension: NEUROCOMPUTING Special Issue on "Pattern Recognition in Graphical Domains" In-Reply-To: <46DEE0D3.7030607@dii.unisi.it> References: <46DEE0D3.7030607@dii.unisi.it> Message-ID: <4725FDCE.1080809@dii.unisi.it> ** Our apologies if you receive multiple copies of this announcement ** Following requests from prospective authors, the deadline for submissions of papers to the NEUROCOMPUTING Special Issue on "Pattern Recognition in Graphical Domains" has been extended to November 12, 2007. Please note that no additional extension can be granted. *********************************************************************** Call for Papers NEUROCOMPUTING Special Issue on PATTERN RECOGNITION IN GRAPHICAL DOMAINS Neurocomputing is seeking original and unpublished manuscripts for a Special Issue on "Pattern Recognition in Graphical Domains", scheduled for publication in June/July 2008. Traditional machine learning applications usually cope with graphs by a preprocessing procedure that transforms structured data to simpler representations. This approach relies on what is called the "feature extraction" process, but it turns out to be quite unnatural for several situations where data are intrinsically organized as graphs, i.e. relationships exist among atomic sub-entities. Unfortunately, valuable information may be lost during the preprocessing and, as a consequence, classical methods may suffer from poor performance and generalization. Therefore, recursive or nested representations, as opposed to "flat" attribute-value data organizations, seem to be more adequate for many relevant problems arising from chemistry, bioinformatics, and the World Wide Web. Recent studies on statistical pattern recognition and neural networks show possible directions to exploit structural information in problems which are inherently of sub-symbolic nature. This special issue is intended to propose a critical re-thinking of the classic learning approaches and to investigate on possible new methodologies and applications of pattern recognition in graphical domains. Submitted articles must not have been previously published and must not be currently submitted for publication elsewhere. Topics of interest include, but are not limited to, the following: - Neural Network Models for Graphs - Support Vector Machines and Kernel Methods for Graphs - Probabilistic Models for Graphs - Statistical Relational Learning - Pattern Recognition Applications Involving Graphical Data Submission procedure: Manuscript should follow the standard guidelines of the Neurocomputing journal. Guidelines for formatting papers can be found in the Guide for Authors at http://www.elsevier.com/wps/find/journaldescription.cws_home/505628/authorinstructions Prospective authors should submit an electronic copy of their complete manuscript through the Elsevier online submission system at http://ees.elsevier.com/neucom/ by November 5, 2007. Important dates: Manuscript submission deadline: ***November 12, 2007*** First notification: January 25, 2008 Revised manuscript submission: February 29, 2008 Notification of final decision: April 11, 2008 Final manuscript due: April 25, 2008 Publication of special issue: June/July 2008 Guest Editors: Monica Bianchini Universit? di Siena Siena, Italy e-mail: monica at dii.unisi.it Franco Scarselli Universit? di Siena Siena, Italy e-mail: franco at dii.unisi.it Information on the Special Issue are also available at http://www.dii.unisi.it/~monica/NeuroSI/ From tgd at eecs.oregonstate.edu Wed Oct 31 01:58:28 2007 From: tgd at eecs.oregonstate.edu (Thomas G. Dietterich) Date: Tue, 30 Oct 2007 22:58:28 -0700 Subject: Connectionists: Faculty Positions in Computer Vision, Natural Language at Oregon State Message-ID: <01d101c81b83$0fbe61e0$800101df@oregone1295e5a> We are hiring in areas relevant to Connectionists. The School of Electrical Engineering and Computer Science at Oregon State University invites applications for up to three tenure-track positions in Computer Science. The School of EECS strongly encourages teamwork and collaboration within the School, and with other departments and universities. We are particularly interested in candidates who can contribute richness and depth to our Graphics/Visualization, End-User Software Engineering and Machine Learning groups. The following areas are strong possibilities for collaboration with these groups: Computer Vision; Human Computer Interaction; Natural Language Processing; Parallel and Distributed Computing (including multi-core and data center computing); Programming Languages; Software Engineering; and Theoretical Computer Science (including algorithms and optimization). Details will be available next week at http://eecs.oregonstate.edu/faculty/openings.php Thomas G. Dietterich Voice: 541-737-5559 School of EECS FAX: 541-737-1300 1148 Kelley Engineering Center URL: http://eecs.oregonstate.edu/~tgd Oregon State University, Corvallis, OR 97331-5501 From julien.mayor at psy.ox.ac.uk Wed Oct 31 06:51:07 2007 From: julien.mayor at psy.ox.ac.uk (Julien Mayor) Date: Wed, 31 Oct 2007 10:51:07 +0000 Subject: Connectionists: 11th Neural Computation and Psychology Workshop (NCPW11) call for abstracts Message-ID: <95F8EC9A-5357-41E2-A008-CC54B0AB3599@psy.ox.ac.uk> ***Apologies for cross-postings*** *** CALL FOR ABSTRACTS *** NCPW11 11th Neural Computation and Psychology Workshop Oxford, UK 16-18 July 2008 http://www.psy.ox.ac.uk/babylab/NCPW/index.html ***** We cordially invite you to participate in the 11th Neural Computation and Psychology Workshop (NCPW11), to be held at University of Oxford, from Wednesday 16th to Friday 18th July 2008. 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. There will be no specific theme, but papers must be about emergent models -- frequently, but not necessarily -- of the connectionist/neural network genre, 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: January 15th, 2008 Abstract submission is now open. For more information see the conference website at: http://www.psy.ox.ac.uk/babylab/NCPW/index.html Notification of acceptance: Approx. 15th February 2008. ***** Dates & Deadlines Abstract submission: 15 January 2008 Notification of acceptance: 15 February 2008 Conference: 16-18 July 2008 --------------------------------------------- Dr Julien Mayor Dept of Experimental Psychology University of Oxford OX1 3UD # +44-1865-271400 Email: julien.mayor at psy.ox.ac.uk http://www.psy.ox.ac.uk/babylab/personal/julien.html From jbednar at inf.ed.ac.uk Wed Oct 31 09:15:25 2007 From: jbednar at inf.ed.ac.uk (James A. Bednar) Date: Wed, 31 Oct 2007 13:15:25 +0000 Subject: Connectionists: New Thesis -- Self-Organising Barrel Cortex Message-ID: <18216.32749.814577.766831@lodestar.inf.ed.ac.uk> Message from Stuart Wilson: Dear Colleagues, It is my pleasure to announce the availability of my MSc thesis, completed in August in the School of Informatics at the University of Edinburgh, UK, under the supervision of Dr. James A. Bednar: Self-Organisation Can Explain the Mapping of Angular Whisker Deflections in the Barrel Cortex A topographic mapping of angular whisker deflections has recently been discovered in the barrel cortex of rats (Andermann & Moore, 2006). Characteristics of this map suggest that it could emerge in post-natal development, through self-organisation under a Hebbian learning regime. This hypothesis was tested in a self-organising computational model, from which remarkably similar mappings emerge when whisker deflections are correlated during training. The model is also used to predict the disorganised mappings that might emerge from the real system when whiskers are uncorrelated, random or anti-correlated with one-another during development. URL: http://homepages.inf.ed.ac.uk/jbednar/papers/IM070505.pdf Software: (freely available; includes the complete model) http://www.topographica.org Keywords: barrel cortex, whisker, topographic map, self-organisation, self-organization, modeling, nature-nurture, development Stuart Wilson Prospective PhD student, ABRG, the University of Sheffield & ANC, the University of Edinburgh I will be attending `Barrels XX' (satellite to SfN in San Diego) this weekend to present a poster of this work, and look forward to the opportunity to discuss the ideas of the project with fellow attendees who may be interested. Please contact me if you have any questions or ideas relating to this topic. Stuart From Dave_Touretzky at cs.cmu.edu Wed Oct 31 19:31:07 2007 From: Dave_Touretzky at cs.cmu.edu (Dave_Touretzky@cs.cmu.edu) Date: Wed, 31 Oct 2007 19:31:07 -0400 Subject: Connectionists: PhD Program in Neural Computation at Carnegie Mellon / University of Pittsburgh Message-ID: <9322.1193873467@ammon.boltz.cs.cmu.edu> The Graduate Program in Neural Computation (PNC) at Carnegie Mellon University and the University of Pittsburgh is now accepting applications. In recognition of the increased demand for computationally-oriented researchers, Carnegie Mellon University, in collaboration with the University of Pittsburgh, last year created this Ph.D. program in computational neuroscience. This program is seeking qualified applicants to begin their graduate training in the Fall of 2008. As neuroscientists have applied new technologies to acquire and analyze large data sets, and have developed new models for understanding increasingly complicated neurobiological systems, quantitative methods have become centrally important to their effort. The new graduate program takes advantage of the unusually large and highly collaborative group of faculty and students in neuroscience in the Pittsburgh community, and builds on the existing, but non-degree granting, graduate program of the Center for the Neural Basis of Cognition. Training in all areas of computational neuroscience is available with special focus on application of dynamical systems, machine learning and statistical approaches to the understanding of the brain at all levels. Details about program curriculum, training faculty and contact information are available at: http://www.cnbc.cmu.edu/GradTrain/pnc_index.shtml The online application is available at: https://applyweb.cs.cmu.edu/apply/index.php?domain=11 The deadline for applications is January 1, 2008. From mpardowi at techfak.uni-bielefeld.de Wed Oct 31 08:52:11 2007 From: mpardowi at techfak.uni-bielefeld.de (Michael Pardowitz) Date: Wed, 31 Oct 2007 13:52:11 +0100 Subject: Connectionists: Scholarships for PhD students and Postdocs at CoR-Lab / Bielefeld Message-ID: ;;Apologies for multiple postings The CoR-Lab has been established at Bielefeld University, Germany, as a research centre for intelligent systems and human-machine interaction. The CoR-Lab forms a strategic partnership between Bielefeld University and the Honda Research Institute Europe GmbH, Germany. It will pursue fundamental research in the field of cognitive robots and intelligent systems, where the Honda humanoid robot ASIMO is available as an advanced technological platform. A particular focus of the CoR-Lab will be the interdisciplinary integration of expertise in engineering, computer science, brain science, and cognitive sciences, including the humanities and social sciences. The Graduate School that is associated with the CoR-Lab provides an exciting and stimulating environment for enthusiastic students and creative postdocs, allowing them to pursue research in international teams in close collaboration with an industrial research institute. We invite applications from students and researchers who hold an academic degree (MSc/Diploma/Ph.D.) and share our dreams. Fluency in English is required. A complete application should include certificates and transcripts of records of the completed course of studies, a CV, a cover letter providing information about the qualification and the motivation to do research in the Graduate School, as well as a short description of the research interests with regard to the research areas of the Graduate School. For more information please see: http://www.cor-lab.de/graduate_school Please send your application until 30 November 2007 (preferably in PDF format) to the Managing Director of the Graduate School: Bielefeld University CoR-Lab Graduate School Dr. Carola Haumann 33594 Bielefeld Germany email: chaumann at cor-lab.uni-bielefeld.de ? From wduch at is.umk.pl Wed Oct 31 05:14:45 2007 From: wduch at is.umk.pl (Wlodzislaw Duch) Date: Wed, 31 Oct 2007 10:14:45 +0100 Subject: Connectionists: Neuroengineering positions Message-ID: <000d01c81b9e$7c391e80$74ab5b80$@umk.pl> Fatronik, a well funded private company in San Sebastian, Spain, seeks experts to work in the new neuroengineering unit on hi-tech support for people with various disabilities, neurodegenerative diseases (Alzheimer, Parkinson), brain damage, developmental disorders and other biomedical engineering problems. The working language of the research program is English. Ideal candidate for the unit leader should have Ph.D. in biomedical engineering or related subjects (computational neuroscience, mathematical modeling, control of human movement), an excellent CV, ability to manage 10-15 researchers, be well known in the field and have a longer experience in similar positions (>10 years). Candidates with lower experience are encouraged to apply for other positions in this unit. Please mention European Neural Networks Society (ENNS) if you apply for any of these positions. Please see "jobs" section at the ENNS web page: http://www.e-nns.org/ As these are Wiki pages you may add more job positions there yourself. Regards, Wlodek Duch ________________________________ Google: W. Duch From raphael.ritz at incf.org Wed Oct 31 12:07:10 2007 From: raphael.ritz at incf.org (Raphael Ritz) Date: Wed, 31 Oct 2007 17:07:10 +0100 Subject: Connectionists: INCF Neuroinformatics demos at SfN Message-ID: <4728A82E.50803@incf.org> The International Neuroinformatics Coordinating Facility (INCF; http://incf.org) is organizing live demonstrations of present databases, tools for visualization and analysis of neuroscience data, and environments for modeling and simulation of nervous system functions at this years Society for Neuroscience meeting in San Diego, November 3-7. The demos take place in the exhibit area at booth #4924 (this is close to the NIH booth). Detailed program and abstracts are available from http://incf.org Raphael From sylvain.chartier at uottawa.ca Wed Oct 31 12:25:27 2007 From: sylvain.chartier at uottawa.ca (Sylvain Chartier) Date: Wed, 31 Oct 2007 12:25:27 -0400 Subject: Connectionists: PhD positions announcement - Recurrent Associative Memories for Categorization and Classification Message-ID: <3F5C7E60AF962A4C87615893F1A4310B02860E11@MSMAIL2.uottawa.o.univ> Two PhD positions in Psychology are available to work with Dr. Sylvain Chartier at the new COmputational NEurodynamics Laboratory (CONEL) at the University of Ottawa. The CONEL laboratory is part of the School of Psychology (http://www.socialsciences.uottawa.ca/psy/eng/index.asp) and the Center for Neural Dynamics (http://www.neurodynamic.uottawa.ca/). The CONEL aims to better understand how human cognitive system accomplishes the complex task of create (and enhance) representations from patterns as well as recognize, identify, categorize and classify them using recurrent associative memories. Further details about the laboratory's research activities can be found at http://aix1.uottawa.ca/~schartie/ The ideal candidates should have programming experience with interpreted language (Matlab and/or Mathematica) and should have the mathematical basis to perform computational modeling. Applicants should complete the application package that can be found at http://www.socialsciences.uottawa.ca/psy/eng/prog2_future_students.asp?ID=adm and send an email to Dr. Chartier (sylvain.chartier at uottawa.ca) about their intention to apply. Deadline: January 3rd, 2008.