From terry at salk.edu Mon Feb 4 19:14:54 2008 From: terry at salk.edu (Terry Sejnowski) Date: Mon, 04 Feb 2008 16:14:54 -0800 Subject: Connectionists: 2007 NIPS Tutorials In-Reply-To: Message-ID: NIPS 2007 Tutorials can be downloaded from: http://nips.cc/Conferences/2007/Program/schedule.php?Session=Tutorials Michael Lewicki Sensory Coding and Hierarchical Representations/ Robert Schapire Theory and Applications of Boosting Leon Bottou, Andrew Moore, and Alex Gray Learning Using Many Examples Tomaso Poggio Visual Recognition in Primates and Machines Geoffrey Hinton Deep Belief Nets Ben Taskar Structured Prediction Click the video icons on the right and the slide icons for pdf slides. All the tutorial videos are 2 X one hour except "Structured Prediction" which is a single 2 hour video. Terry ----- From hiro at brain.riken.jp Sat Feb 2 01:54:18 2008 From: hiro at brain.riken.jp (hiroyuki nakahara) Date: Sat, 02 Feb 2008 15:54:18 +0900 Subject: Connectionists: FYI RIKEN BSI Summer School 2008 Message-ID: <20080202155343.ECAE.HIRO@brain.riken.jp> Dear colleagues, I forward the following information. - Hiro Nakahara Call for Applications RIKEN Brain Science Institute 2008 Summer Program URL: http://www.brain.riken.jp/en/summer/index.html Application deadline; February 29, 2008 Lecture Course: July 28 - August 8 Internship: June 25 - August 20 Every summer, RIKEN Brain Science Institute (Wako, Japan; nearby Tokyo) organizes the international summer school. The short summary of the summer school this year is attached below. Application guideline, further information of the coming and past summer schools and etc are on the above URL. The summer school program has two types of applications. One ("Lecture Course application") is to attend the lecture course. The other ("Inernship application") is to have a two-months intership in one of laboratories at RIKEN Brain Science Instiute in addition to the attendance at the lecture course. In both types, financial support for travel and accommodation will be considered for those without external funding. For questions, please send any general inquiry not to me but to the adminstration or organizing committee of the summer school shown in the above summer school URL, while you can send inquiry to me if it is specifically about my laboratory (lab URL: http://www.itn.brain.riken.jp). 2008 Summer Program Developmental Foundations of Brain Function and Dysfunction Advances in our understanding of the mechanisms of neural development have made significant impacts on broader areas of brain science. These areas include studies of neural plasticity and neuropsychiatric disorders, and improvement of therapeutic strategies for nervous system damage. This year, the RIKEN Brain Science Institute has organized a series of lectures by leading scientists that emphasize the links between the molecular and cellular bases of neural development, plasticity, dysfunctions and repair. 2008 Summer Program Lecture Schedule [Invited Lecturers] Silvia BUNGE (UC Berkeley) Eric COURCHESNE (UC San Diego) Marie FILBIN (Hunter College, CUNY) Lawrence GOLDSTEIN (UC San Diego) Elizabeth GOULD (Princeton Univ.) Masanobu KANO (Tokyo Univ.) Pat LEVITT (Vanderbilt Univ.) Fumio MATSUZAKI (RIKEN CDB) Randolph NUDO (Kansas Univ.) Hideyuki OKANO (Keio Univ.) Hitoshi SAKANO (Tokyo Univ.) Akira SAWA (Johns Hopkins Univ.) Stephen SMITH (Stanford Univ.) Lawrence ZIPURSKY (UC Los Angeles) [Special lecturer] Sandra BLAKESLEE (Science writer) [BSI Speakers] Teiichi FURUICHI (Lab. for Molecular Neurogenesis) Atsushi IRIKI (Lab. for Symbolic Cognitive Development) Hiroyuki KAMIGUCHI (Lab. for Neuronal Growth Mechanisms) Hitoshi OKAMOTO (Lab. for Developmental Gene Regulation) Keiji TANAKA (Lab. for Cognitive Brain Mapping) Kazuhiro YAMAKAWA (Lab. for Neurogenetics) --------------------------------- Hiro Nakahara Lab for Integrated Theoretical Neuroscience RIKEN Brain Science Institute 2-1 Hirosawa Wako Saitama, 351-0198, Japan Email: hn at brain.riken.jp Lab webpage: http://www.itn.brain.riken.jp/ From ASIM.ROY at asu.edu Sun Feb 3 14:11:00 2008 From: ASIM.ROY at asu.edu (Asim Roy) Date: Sun, 3 Feb 2008 12:11:00 -0700 Subject: Connectionists: New Paper: "Connectionism, controllers and a brain theory" - Paper now available on my website Message-ID: There are many requests for this paper. So it is posted on my website now. Here's the link: http://wpcarey.asu.edu/pubs/index.cfm?fct=details&article_cobid=2216410&author_cobid=1039524&journal_cobid=2216411 Feel free to contact me if there are questions. Asim Roy ================================================================================================= (My apologies if you get multiple copies of this.) The following paper has been accepted for publication in IEEE Transactions on Systems, Man and Cybernetics, Part A. It essentially invalidates significant parts of connectionism. There were numerous public debates over the past decade on connectionism, some at international neural net conferences and some online through various mailing list. Some of the debate summaries are available at the CompNeuro archive site and listed below. TITLE: Connectionism, controllers and a brain theory AUTHOR: Asim Roy, Arizona State University, Tempe, AZ 85287-4606, USA ABSTRACT: This paper proposes a new theory for the internal mechanisms of the brain. It postulates that there are controllers in the brain and that there are parts of the brain that control other parts. Thus the theory refutes the connectionist theory that there are no separate controllers in the brain for higher-level functions and that all control is "local and distributed" at the level of the cells. Connectionist algorithms themselves are used to prove this theory. Plus there is evidence in the neuroscience literature to support this theory. Thus the paper proposes a control theoretic approach to understanding how the brain works and learns. That means that control theoretic principles should be applicable to developing systems similar to the brain. =============================================================================================================== For those interested, summaries of prior debates on the basic ideas of connectionism are available at the CompNeuro archive site. Here is a partial list of the debate summaries that are available . - Some more questions in the search for sources of control in the brain - BRAINS INTERNAL MECHANISMS - THE NEED FOR A NEW PARADIGM - COULD THERE BE REAL-TIME, INSTANTANEOUS LEARNING IN THE BRAIN? - CONNECTIONIST LEARNING: IS IT TIME TO RECONSIDER THE FOUNDATIONS? - DOES PLASTICITY IMPLY LOCAL LEARNING? AND OTHER QUESTIONS - Connectionist Learning - Some New Ideas/Questions =================================================================================================== From esann at dice.ucl.ac.be Thu Feb 7 02:22:42 2008 From: esann at dice.ucl.ac.be (esann@dice.ucl.ac.be) Date: Thu, 7 Feb 2008 08:22:42 +0100 Subject: Connectionists: ESANN'2008 programme Message-ID: <001301c8695a$3a6db2e0$43ed6882@maxwell.local> (Our apologies if you get multiple copies of this message, despite our precautions) ============================================= ESANN'2008 16th European Symposium on Artificial Neural Networks Advances in Computational Intelligence and Learning Bruges (Belgium) - April 23-24-25, 2008 Preliminary program ============================================= The preliminary program of the ESANN'2008 conference is now available on the Web: http://www.dice.ucl.ac.be/esann For those of you who maintain WWW pages including lists of related machine learning and artificial neural networks sites: we would appreciate if you could add the above URL to your list; thank you very much! For 16 years the ESANN conference has become a major event in the field of neural computation and machine learning. ESANN is a selective conference focusing on fundamental aspects of artificial neural networks, machine learning, statistical information processing and computational intelligence. Mathematical foundations, algorithms and tools, and applications are covered. This year, around 100 scientific communications will be presented, covering most areas of the neural computation and related fields. The program of the conference can be found at http://www.dice.ucl.ac.be/esann, together with practical information about the conference venue, registration, etc. Other information can be obtained by sending an e-mail to esann at dice.ucl.ac.be. ======================================================== ESANN - European Symposium on Artificial Neural Networks http://www.dice.ucl.ac.be/esann * For submissions of papers, reviews,... Michel Verleysen Univ. Cath. de Louvain - 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 lfranco at lcc.uma.es Thu Feb 7 10:10:47 2008 From: lfranco at lcc.uma.es (lfranco@lcc.uma.es) Date: Thu, 07 Feb 2008 16:10:47 +0100 Subject: Connectionists: cfp: special session on constructive neural networks, ICANN 2008 Message-ID: <200802071510.m17FAlPb023679@sol10.lcc.uma.es> Dear Colleague, We would like to invite you to submit a paper to a special session on "CONSTRUCTIVE NEURAL NETWORKS" that we are organizing and that will take place at the 18th ICANN 2008, Prague, September 3-6, 2008. The ICANN ? International Conference on Artificial Neural Networks is an annual conference organized since 1998 by the ENNS - European Neural Network Society in co-operation with the International Neural Network Society and the Japanese Neural Network Society, and is a premier event in all topics related to neural networks. The special session, aimed to foster the interaction between researchers in the area, will include speakers from neural networks, discrete mathematics, machine learning, etc. The idea is to present and discuss, in a friendly atmosphere, recent advances in the construction and application of constructive methods for neural networks. Contributed papers are due on March 3, 2008. If you are thinking on submitting your work, consider contacting us in case you have any doubt or would want any further details. You can find information about ICANN'08 and in the following web address: http://www.ICANN2008.org/ See also the call for the special session at the end of this e-mail. Please, feel free to send this message to your colleagues working on the topic of the session and related areas. Thank you very much and we look forward to your participation. Best wishes, David Elizondo, Leonardo Franco and Jose Jerez Session organizers ICANN'08 ================================================== ICANN 2008 International Conference on Artificial Neural Networks September 3-6, 2008 http://www.icann2008.org/ Prague, Czech Republic CALL FOR PAPERS - SPECIAL SESSION "CONSTRUCTIVE NEURAL NETWORKS" organized by David A. Elizondo Centre for Computational Intelligence School of Computing De Montfort University Leicester, UK Tel: +44(0) 116-255-1551, Fax: +44(0) 116-207-8159 email: elizondo at dmu.ac.uk Leonardo Franco Dept. of Computer Science University of Malaga Malaga, Spain Tel: +34(0) 952-133304, Fax: +34(0) 952-131397 email: lfranco at lcc.uma.es Web: http://www.lcc.uma.es/~lfranco/ Jos? M. Jerez Dept. of Computer Science University of Malaga Malaga, Spain Tel: +34(0) 952-132859, Fax: +34(0) 952-131397 email: jja at lcc.uma.es Purpose of the session This special session will be devoted to constructive neural networks and other incremental learning algorithms that constitute an alternative to the standard method using backpropagation. These algorithms provide an incremental way of building neural networks with reduced topologies for classification problems. Furthermore, these neural networks produce a multilayer topology, which together with the weights, are determined automatically by the constructing algorithm and thus avoid the search for finding a proper neural network architecture. Another advantage of these algorithms is that convergence is guaranteed by the method. A growing amount of current research in neural networks is oriented towards this important topic. Providing constructive methods for building neural networks can potentially create more compact models which can easily be implemented in hardware and used on embedded systems. The purpose of this session is to gather together some of the leading investigators and research groups in this growing area, and to provide an overview of the most recent advances on the techniques being developed for constructive neural networks and their applications. Other topics of interest include methods for estimating what neural architectures are best suited for a given problem, comparisons of different network topologies and genetic or evolving methods for architecture design. From valenti at dsi.unimi.it Wed Feb 6 13:12:30 2008 From: valenti at dsi.unimi.it (Giorgio Valentini) Date: Wed, 06 Feb 2008 19:12:30 +0100 Subject: Connectionists: CFP: SUEMA 2008 Message-ID: <47A9F88E.1090808@dsi.unimi.it> ================================================================ CALL FOR PAPERS Workhop on Supervised and Unsupervised Ensemble Methods and Their Applications - SUEMA 2008 http://suema08.dsi.unimi.it European Conference on Artificial Intelligence - ECAI 2008 Patras (Greece) 21-22 July 2008 ================================================================= Dear colleague, we are pleased to invite you to submit a paper to the workshop Supervised and Unsupervised Ensemble Methods and Their Applications (SUEMA 2008), that will be held in the context of the European Conference on Artificial Intelligence (ECAI 2008) in Patras. This second edition, organized within the PASCAL2 (Pattern Analysis, Statistical Modelling and Computational Learning) European Network of Excellence, intends to provide a forum for researchers in the field of Artificial Intelligence and Machine Learning to discuss topics related to ensemble methods and their applications. New trends and challenges in the theory and methods of ensembles of learning machines will be discussed, with a particular focus on the application of principled ensemble methods to emerging real-life problems. More information about the topics of the workshop are available at the workshop web-site: http://suema08.dsi.unimi.it With best regards Oleg Okun and Giorgio Valentini. --- IMPORTANT DATES Submission 7th May 2008 Notification 26th May 2008 Camera Ready 5th June 2008 --- Submission of papers The style for SUEMA 2008 papers is the same of the main conference ECAI 2008. The maximum length is 5 pages (double columns). For Latex users the necessary files can be downloaded from: http://www.ece.upatras.gr/ecai2008/documents/ecai2008.zip. The authors should submit the papers by e-mail to the workshop chairs Oleg Okun (oleg at ee.oulu.fi) and Giorgio Valentini (valentini at dsi.unimi.it). --- Workshop proceedings Each workshop participant will receive a hardcopy of the proceedings of the workshop (s)he participates, as well as a CD with the proceedings of all ECAI'08 workshops. We are managing to publish workshop extended papers in an edited book, as for the 1st SUEMA workshop (that will be published by Springer-Verlag). --- Registration Partcipants to the workshop should register directly to the main ECAI 2008 conference (workshops' early registration deadline is set to 16 June 2008) ------ Main topics The main topics of the conference include (but not are limited to): New ensemble methods raised from new real world supervised and unsupervised learning problems Application of ensemble methods in various branches of science and technology: bioinformatics, medical informatics, computer security, economics, ecology, meteorology and weather forecast, image analysis and signal processing, satellite image analysis. Fusion of multiple-source/multi-sensor data Unsupervised ensemble methods for discovering structures in unlabeled real data Unsupervised ensemble approaches to assess the reliability/validity of clusters discovered in real data Combination techniques and methods to generate multiple base learners from different features and data Dynamic member selection for including into an ensemble Heterogeneous ensembles of base learners Variants of re-sampling-based methods (bagging, boosting) Ensemble methods for supervised multi-class classification and regression Supervised and unsupervised ensemble methods for structured domains --- SUEMA Scientific Program Committee Nicolo' Cesa-Bianchi, University of Milano, Italy Carlotta Domeniconi, George Mason University, USA Robert Duin, Delft University of Technology, the Netherlands Mark Embrechts, Rensselaer Polytechnic Institute, USA Ana Fred, Technical University of Lisboa, Portugal Joao Gama, University of Porto, Portugal Giorgio Giacinto, University of Cagliari, Italy Larry Hall, University of South Florida, USA Ludmila Kuncheva, University of Wales, UK Francesco Masulli, University of Genova, Italy Petia Radeva, Autonomous University of Barcelona, Spain Juan Jose' Rodriguez, University of Burgos, Spain Fabio Roli, University of Cagliari, Italy Carlo Sansone, Federico II University of Napoli, Italy Jose' Salvador Sanchez, University Jaume I, Spain Grigorios Tsoumakas, Aristotle University of Thessaloniki, Greece Jordi Vitria', Autonomous University of Barcelona, Spain Ioannis Vlahavas, Aristotle University of Thessaloniki, Greece Terry Windeatt, University of Surrey, UK --- Workshop Chairs Oleg Okun, Dept. of Electrical and Information Engineering, University of Oulu, Finland oleg at ee.oulu.fi Giorgio Valentini, DSI, Dept. of Computer Science University of Milano, Italy valentini at dsi.unimi.it -- =========================================== Giorgio Valentini D.S.I. - Universita' degli Studi di Milano - Italia Phone: +39 (02) 503.16225 e-mail: valentini at dsi.unimi.it http://homes.dsi.unimi.it/~valenti =========================================== -- =========================================== Giorgio Valentini D.S.I. - Universita' degli Studi di Milano - Italia Phone: +39 (02) 503.16225 e-mail: valentini at dsi.unimi.it http://homes.dsi.unimi.it/~valenti =========================================== From gbrown at cs.man.ac.uk Thu Feb 7 06:18:52 2008 From: gbrown at cs.man.ac.uk (Gavin Brown) Date: Thu, 07 Feb 2008 11:18:52 +0000 Subject: Connectionists: Research Associate in Machine Learning (Bristol, UK) Message-ID: <47AAE91C.6090307@cs.man.ac.uk> Please see below for an open position at Bristol... ------------------------------------------------------------------- Research Associate Department of Computer Science You will act as RA on the Adaptive Dynamic Ensemble Prediction Techniques (ADEPT) project which aims to merge theory and practice between two sub-fields of Machine Learning: Ensemble methods and evolutionary Learning Classifier Systems. Though they have developed largely in isolation we believe these areas are both significantly overlapping and complementary. We aim to unify the two areas and use the best of both to develop novel predictor-independent methods applicable to a range of machine learning problems. This is an exciting opportunity to conduct original scientific research at the cutting edge of machine learning from within a large and diverse group. You will deliver empirical results, software, theoretical analyses, and oral and written reports. You will be based with the department of computer science's Machine Learning and Biological Computation group http://www.cs.bris.ac.uk/Research/MachineLearning/ which is part of Intelligent Systems at Bristol http://intelligentsystems.bristol.ac.uk/ and home to a world class MSc Machine Learning and Data Mining degree http://www.cs.bris.ac.uk/handbook/2007-8/course-msc-cs.html You will collaborate with other project members in Bristol, Manchester and Birmingham, and generally contribute to the academic life of the department. Objectives of the project include: - Developing ensemble methods for dynamic / non-stationary problems - Developing ensemble methods for multi-step problems - Building a statistical framework for learning classifier systems by drawing on the pattern recognition literature - Adapting the explicit probabilistic modelling approach of Estimation of Distribution Algorithms (EDAs) to our novel methods to better solve problems with interactions between features Essential qualifications: - A PhD in a relevant area - An excellent track record of publishing high-quality research - Excellent communication skills (oral and written) - Initiative and independence - A strong background in machine learning, pattern recognition and/or evolutionary computation. Desirable qualifications: - Evidence of the ability to carry out research independently and to a high standard - A background in one or more of the following: - evolutionary computation - reinforcement learning - ensemble methods - multi-objective learning methods - probabilistic modelling approaches - evolutionary learning classifier systems - Good software engineering skills (including e.g. familiarity with UML and design patterns and experience with substantial projects) Salary up 28,290 UKP depending on qualifications and experience Timescale of appointment: Contract: Fixed Term Contract 35 months from start date Anticipated interview date: 03 March 2008 Anticipated start date: 01 April 2008 Closing date for applications: 9:00am on 18 February 2008 Contact Tim Kovacs for informal enquiries: http:/www.cs.bris.ac.uk/~kovacs For further information including application procedures see http://www.bris.ac.uk/boris/jobs/ads?ID=71287 ---------------------------------------------------------------- From stefan.wermter at sunderland.ac.uk Fri Feb 8 10:25:18 2008 From: stefan.wermter at sunderland.ac.uk (Stefan Wermter) Date: Fri, 08 Feb 2008 15:25:18 +0000 Subject: Connectionists: Multimodal communication: NN special issue call for papers Message-ID: <47AC745E.8000303@sunderland.ac.uk> ------------------------------------------------------------------ CALL FOR PAPERS ?Multimodal communication and what it means to communicate? Special Issue of Neural Networks The general aim of this special issue is to focus on ?What it means to communicate? and to understand the neural, cognitive, formal, computational and developmental features that have led to communication differences between humans and animals. A number of interesting and successful research directions have been explored including learning by imitation, examining the origin of human rule based reasoning, studying the neural origins of language, exploring the evolutionary origins of the human mind, researching into verbal and nonverbal communication, using and interpreting signs, characterising human language by structural complexity, and representing abstract concepts. To complement this, computation and neural robotics aims to explore a multimodal understanding of verbal and visual communication for embodied action understanding leading towards formal models of knowledge representation and reasoning. We are looking for submissions in the areas including * Neural network models and theories for communication * Knowledge representation and hybrid models for communication * Plausible reasoning and uncertainty in communication * Multimodal integration and crossmodal neural models * Formals models of human communication * The role of neurally grounded emotion in communication * Bioinspired communicating agents with reasoning * Neuroscience-inspired robots and communication * Communicating agents with an understanding of emotion * Developmental approaches to communication * Neural networks and imitation * Neural approaches to sign and gesture * Mirror neuron system and multimodal integration * Embodiment of communication and action in robots * Language and action circuits * The role of formal memory in communication Co-Editors: Stefan Wermter (University of Sunderland) Vittorio Gallese (University of Parma) Friedemann Pulvermuller (MRC, Cambridge) John Taylor (King?s College, London) Submission: Deadline for submission: May1, 2008 Notification of acceptance: August1, 2008 Deadline for submission of revised papers: October1, 2008 Notification of final acceptance: November1, 2008 Format: see papers for the journal Neural Networks. http://www.elsevier.com/wps/find/journaldescription.cws_home/841/description Submissions: To be sent to: Stefan Wermter as pdf files at the address below *************************************** Professor Stefan Wermter Centre for Hybrid Intelligent Systems School of Computing and Technology University of Sunderland St Peters Way Sunderland SR6 0DD United Kingdom phone: +44 191 515 3279 fax: +44 191 515 3553 email: stefan.wermter AT sunderland.ac.uk http://www.his.sunderland.ac.uk/~cs0stw/ http://www.his.sunderland.ac.uk/ **************************************** From xuedong.huang at gmail.com Sat Feb 9 06:56:09 2008 From: xuedong.huang at gmail.com (XueDong Huang) Date: Sat, 09 Feb 2008 11:56:09 -0000 Subject: Connectionists: New Paper: "Connectionism, controllers and a brain theory" - Paper now available on my website In-Reply-To: References: Message-ID: <47AD94D6.5050204@gmail.com> Asim Roy wrote: > There are many requests for this paper. So it is posted on my website now. Here's the link: > > http://wpcarey.asu.edu/pubs/index.cfm?fct=details&article_cobid=2216410&author_cobid=1039524&journal_cobid=2216411 > Dear Asim, If I understand you right, you state some limitations in current "connectionnist theories", ones that claim that there are no master or slave in an executive chain or network. In this paper, you propose that there exist some masters that behave without depending on the feedback of the slaves they are controlling. Do you know that "connectionnist theories" or any other simplified theories in human-brain studies are there to support some easy and computerizabled models of how our human brain functions ? When you criticize them, do you have any evidence why your new "theory" should bring a better outcome to current researches on this ? Show us some move proofs Thanks XDH From triesch at fias.uni-frankfurt.de Wed Feb 20 10:29:41 2008 From: triesch at fias.uni-frankfurt.de (Jochen Triesch) Date: Wed, 20 Feb 2008 16:29:41 +0100 Subject: Connectionists: FIAS Summer School Theoretical Neuroscience & Complex Systems (2-24 August 2008, Frankfurt, Germany) Message-ID: <47BC4765.2040803@fias.uni-frankfurt.de> Announcement and Call for Applications: The Frankfurt Institute for Advanced Studies (FIAS) is organizing a PENS-Blackwell Summer School on: Theoretical Neuroscience & Complex Systems Frankfurt, Germany, August 2-24, 2008 We invite applications for this three-week summer school from graduate students and post-docs in experimental and computational neuroscience and theoretical physics. Up to 30 students will be accepted. The school is supported by PENS - the Programme of European Neuroscience Schools, an IBRO-FENS collaboration. Application deadline: Monday, April 7. School: http://fias.uni-frankfurt.de/neuro_school/ Application: http://fens.mdc-berlin.de/pens/2008/schools/frankfurt/ Scientific Directors: W. Maass, C. von der Malsburg, G. Pipa, W. Singer, J. Triesch, M. Tsodyks Lecturers and Tutors Include: * Dana Ballard, UT Austin, USA * Emery Brown, Harvard/MIT, Cambridge, USA * Yang Dan, UC Berkeley, USA * Sophie Deneve, CNRS, France * Rodney Douglas, ETH Zurich, Switzerland * Jordi Garcia-Ojalvo, Universitat Politecnica de Catalunya, Spain * Wulfram Gerstner, EPFL, Switzerland * Ann Graybiel, MIT, USA * Mary Hayhoe, UT Austin, USA * Jurgen Kurths, University of Potsdam, Germany, * Jorg Lucke, FIAS, Germany * Wolfgang Maass, Technische Universitat, Graz, Austria * Christoph von der Malsburg, FIAS, Germany * Sergio Neuenschwander, MPIH, Germany * Danko Nikolic, MPIH, Germany * Gordon Pipa, FIAS and MPIH, Germany * Austra Saudargiene, Vytautas Magnus University, Lithuania * Kerstin Schmidt, MPIH, Germany * Wolf Singer, MPIH, Germay * Olaf Sporns, Indiana University, USA * Jochen Triesch, FIAS, Germany * Peter Uhlhaas, MPIH, Germany * Raul Vicente, FIAS and MPIH, Germany * Cornelius Weber FIAS, Germany * Michael Wibral, MPIH, Germany * Kai Willadsen, FIAS, Germany * Junmei Zhu, FIAS, Germany Topics: mathematical and biological foundations of theoretical neuroscience; analysis of neural data; modelling of neurons and small networks; plasticity and learning; modelling higher cognitive functions, embodiment Format: - a 3-day pre-school to establish a common foundation and language - daily lectures by renowned international faculty - hands-on work on self-defined projects in interdisciplinary groups - round table discussions - a social and cultural program Registration Fees, Travel Expenses and Waivers: The school covers the expenses for accommodation and the cultural program. Students will have to pay a registration fee of 400 EURO. A small number of grants are available for students who need support for their travel expenses and a waiver for the registration fee. They are primarily intended for students from disadvantaged regions. If you consider applying for financial support, please indicate this in your application. We also need an estimate of your travel expenses (in EUROs). For further information, please contact: Neuro School Frankfurt Institute for Advanced Studies Johann Wolfgang Goethe University Ruth-Moufang-Str. 1 60438 Frankfurt am Main Germany email: neuro_school at fias.uni-frankfurt.de tel: +49 69 798 47601 fax: +49 69 798 47611 From R.Borisyuk at plymouth.ac.uk Wed Feb 20 11:26:19 2008 From: R.Borisyuk at plymouth.ac.uk (Roman Borisyuk) Date: Wed, 20 Feb 2008 16:26:19 +0000 Subject: Connectionists: From Neuron to Behaviour: MSc programme at University of Plymouth, UK Message-ID: <6CD19ED93A7A8F4593955A11621242C20658F68625@ILS133.uopnet.plymouth.ac.uk> >From Neuron to Behaviour: MSc programme on Neural Systems, Cognitive Neuroscience and Brain Function We are pleased to announce a new MSc programme Theoretical and Computational Neuroscience at University of Plymouth, UK. The course is full time for 12 months, and starts in September 2008. This exciting multidisciplinary programme provides you with a comprehensive knowledge of principles and concepts of brain functioning. It covers anatomy and physiology of brain structures; mechanisms controlling dynamics of neural activity, state-of-the-art brain imaging techniques; the relation between neural activity and cognitive functions such as object and sound recognition, attention, memory; and many other intriguing topics related to understanding how the brain makes sense of the world and how it organises behaviour. The programme is unique in Europe in providing training within the quickly evolving cutting-edge discipline of Theoretical Neuroscience. It is especially suitable for students from the experimental sciences (psychology, biology and medicine). Students with engineering, computing, physics, mathematics and similar degrees wishing to broaden their horizons toward applications in the amazing field of brain modelling will find this programme very attractive. Programme highlights * You will participate in an interdisciplinary, new and unique programme studying fundamental principles of information processing in the brain * You will learn how incoming sensory signals become our perception, how we recognise images and sounds, form attention focus on the most important information, memorise events and their sequences, develop behavioural responses, control locomotion etc. * You will develop confidence and competence in the theoretical and practical dimensions of modelling in neuroscience, through taught modules, practical work, exercises and project development * You will become experienced in analysing experimental data of cognitive and neurophysiological experiments; new medical devices and methods to study the brain and treat neurological disorders; brain inspired algorithms and neuromorphic techniques of finding solutions to complex problems * You will be taught by recognised experts with worldwide links to research organisations and projects * You can take advantage of our partnership with centres of computational neuroscience in the UK, Europe, USA and Japan The programme is developed in the Centre for Theoretical and Computational Neuroscience (CTCN) ( http://www.research.plymouth.ac.uk/ctcn/ ). The CTCN is one of the leading centres in the field of theoretical neuroscience. The Centre has bought together a range of international experts from various backgrounds. This rich mix of computational, mathematical and neuroscience expertise provides a unique opportunity for students to acquire multidisciplinary training. The programme should provide an excellent training for those wishing to pursue an academic research career. In particular, it provides a broad multidisciplinary foundation for students wishing to undertake further research training at PhD level. Also the programme opens exciting perspectives for students wishing to work for the pharmaceutical and medical industries and apply their knowledge of new and exciting discoveries in brain research. For example, biomedical engineering requires experts in brain functioning to deal with developments of new drugs to treat mental and neurological diseases, develop prosthetic devices etc. This programme will be attractive for medical professionals wishing to continue their professional development. For more information and application see http://www.plymouth.ac.uk/courses/postgraduate/taught/3068/MSc+Theoretical+and+Computational+Neuroscience Contact details Postgraduate Admissions Faculty of Science University of Plymouth Drake Circus Plymouth PL4 8AA United Kingdom Tel: +44 (0)1752 233093 E-mail: science at plymouth.ac.uk From Eugene.Izhikevich at nsi.edu Wed Feb 20 12:16:08 2008 From: Eugene.Izhikevich at nsi.edu (Eugene M. Izhikevich) Date: Wed, 20 Feb 2008 09:16:08 -0800 Subject: Connectionists: Workshop "Dynamical Systems in Biology", New York University, April , 12-13, 2008 Message-ID: <47BC6058.3090508@nsi.edu> Workshop: DYNAMICAL SYSTEMS IN BIOLOGY to celebrate 70th birthday of Frank C. Hoppensteadt April 12-13, 2008 New York University, NY Program: http://hoppensteadt.org Organizers: Eugene M. Izhikevich, The Neurosciences Institute, San Diego, CA Ying-Cheng Lai, Arizona State University, Tempe, AZ Fern Hunt, National Institute of Standards and Technology April 12: DYNAMICAL SYSTEMS IN POPULATION BIOLOGY Speakers: S. Antman, B. O'Malley, A. Skorokhod, H. Salehi, B. Stone, A. Hastings, J. Cohen, F. Hunt, J. Burke, W. Jaeger, L. Bunimovich, S. Childress, D. McLaughlin April 13: DYNAMICAL SYSTEMS IN NEUROSCIENCE Speakers: F. Skinner, R. Borisyuk, L.-S. Young, W. Miranker, C. Peskin, H. Carrillo, P. Holmes, J. Rinzel, V. Booth, J. Guckenheimer, Y.-C. Lai, E. Izhikevich, F. C. Hoppensteadt Partial travel support may be available for junior participants, subject to NSF approval. -- Eugene M. Izhikevich, Senior Fellow in Theoretical Neurobiology PhD, Mathematics, http://www.nsi.edu/users/izhikevich The Neurosciences Institute, Editor-in-Chief at scholarpedia.org 10640 John J. Hopkins Drive tel:(858) 626-2063 San Diego, CA, 92121, USA fax:(858) 626-2099 From M.Loog at tudelft.nl Thu Feb 21 04:42:03 2008 From: M.Loog at tudelft.nl (M.Loog@tudelft.nl) Date: Thu, 21 Feb 2008 10:42:03 +0100 Subject: Connectionists: Ph.D. Vacancies in Pattern Recognition at Delft University of Technology Message-ID: <6B525D400973B44AB945EFF5F9F0EA3304FA5C@EWIS22.ewi.local> At this moment we have three possibilities for a new PhD project : - Similarity Based Pattern Analysis - Multi-feature Image Analysis - Learning in Multi-Scale Image Analysis All projects are fundamental in the sense that there is no main application involved. All of them, however, fit in other, sometimes more applied projects we are currently participating in. A description is available at http://ict.ewi.tudelft.nl/~duin/PR_PhD_projects_TUDelft.pdf . The projects will be supervised by Dr. Robert P.W. Duin and / or Dr. Marco Loog. Promoter will be Prof.Dr. M.J.T. Reinders. For reactions and information (before April 1, 2008) : M.Loog at TUDelft.nl From m.usher at psychology.bbk.ac.uk Thu Feb 21 12:06:19 2008 From: m.usher at psychology.bbk.ac.uk (DR M USHER) Date: Thu, 21 Feb 2008 17:06:19 +0000 (GMT) Subject: Connectionists: MSc program in Cognitive-and-Decision sciences Message-ID: <20080221170619.ADN87043@mpmail2.bbk.ac.uk> Dear Colleagues, Please forward this advert to any students who might be interested in applying for postgraduate studies (MSc in Cognitive and Decision Sciences) in London at Birkbeck and UCL. Thank you! Marius Usher, | Professor of Cognitive Psychology | Email:m.usher at bbk.ac.uk Birkbeck, University of London | Tel: 44 (0)20 7631 6201 Malet Str., WC1E 7HX, London, UK | Fax: 44 (0)20 7631 6312 --------------------------------------------------------------- MSc Cognitive and Decision Sciences This program studies the cognitive processes underlying human thought and decision making. Key topics include: probabilistic models of the mind; the nature of computational explanation; general principles of cognition; the scope of rational choice models; learning and memory; applications to economics and business. It draws on an outstanding faculty at Birkbeck and UCL, including internationally renowned researchers in psychology, computational modeling, neuroscience and economics. It also takes full advantage of London?s unique position as a global ?hot-spot? for research in cognition, decision-making, and neuroscience, with a high density of research seminars and scientific meetings. The course is suitable for students from a wide range of disciplines, including psychology, economics, neuroscience, philosophy, computer science and statistics. Further information is available on: http://www.bbk.ac.uk/psyc/prospective/pgtaught/msc-cogdes (MSc in Cognitive and Decision Sciences (CoDeS) http://www.decisionsciencenews.com/?p=203 For informal inquiries contact Marius Usher (m.usher @ psychology.bbk.ac.uk) or David Lagnado (d.lagnado @ ucl.ac.uk). Admin. inquiries to: j.vallerine at psychology.bbk.ac.uk or p.horbacki at ucl.ac.uk From d.g.heinke at bham.ac.uk Wed Feb 13 11:52:51 2008 From: d.g.heinke at bham.ac.uk (Dietmar Heinke) Date: Wed, 13 Feb 2008 16:52:51 -0000 Subject: Connectionists: Master in Cognition and Computational Neuroscience Message-ID: <47B3203F.1070002@bham.ac.uk> The School of Psychology, University of Birmingham (UK) offers a Master of Research (M.Res) in Cognition and Computational Neuroscience. This master programme is open to graduate students with a good first degree in Biology, Computer Science, or Physics. The course is ideal for people who are thinking of pursuing a PhD in psychology and want to test the water. For further information see http://www.comp-psych.bham.ac.uk/mres_comp.htm. Five tuition-fee only studentships are available. Deadline for applications is end of July. M.Res degrees in other specializations are available as well (http://psg275.bham.ac.uk/prospect_students_03/postgraduate.htm#masters). For any further enquiries, please contact Mrs Parveen Chahal Tel: +44-121-41-44906 email: p.k.chahal at bham.ac.uk From dorr at inb.uni-luebeck.de Fri Feb 15 08:16:03 2008 From: dorr at inb.uni-luebeck.de (Michael Dorr) Date: Fri, 15 Feb 2008 13:16:03 -0000 Subject: Connectionists: Assistant Professor "Machine Learning in Medicine and Life Sciences", University of Luebeck Message-ID: <200802151415.17882.dorr@inb.uni-luebeck.de> Applications are invited for a position becoming available at the Institute for Neuro- and Bioinformatics, University of Luebeck (Director: Prof. Thomas Martinetz) in the context of the Graduate School "Computing in Medicine and Life Sciences" (http://www.gradschool.uni-luebeck.de) Deadline for applications: March 28, 2008 Starting date: as soon as possible Assistant Professor (Juniorprofessur W1) for Machine Learning in Medicine and Life Sciences The initial appointment will be for 3 years, but can be extended up to 6 years. Responsibilities include conducting independent research, teaching courses relevant to "Machine Learning in Medicine and Life Sciences" (4 hours/week, German and English), and the attraction of extramural funding. The successful candidate is expected to set up their own research group within the Institute for Neuro- and Bioinformatics (http://www.inb.uni-luebeck.de). Applicants must hold a PhD degree and should have experience, as indicated by their academic record, in one or several of the following topics: - machine learning in medical imaging and image processing - machine learning in bioinformatics and biomedical data analysis - theory and novel methods of machine learning The university strongly encourages teamwork and interdisciplinary collaboration, in particular with the recently established Research Centre "Computer Imaging and Computer Aided Diagnosis (CICAD)" and with clinical institutes. The university seeks to increase the number of women in those areas where they are underrepresented and therefore explicitly encourages women to apply. Applications from disabled individuals are especially welcome. Applicants are invited to send their complete resum? (including a list of publications and the names of three possible referees), before March 28, 2008, to: Dekanat der Technisch-Naturwissenschaftlichen Fakult?t Universitaet zu Luebeck Ratzeburger Allee 160 23538 Luebeck Germany or electronically to martinetz at inb.uni-luebeck.de . From friedhelm.schwenker at uni-ulm.de Tue Feb 19 06:14:14 2008 From: friedhelm.schwenker at uni-ulm.de (Friedhelm Schwenker) Date: Tue, 19 Feb 2008 12:14:14 +0100 Subject: Connectionists: Job posting Open Position Message-ID: <47BABA06.6000409@uni-ulm.de> Job posting Open position --------------------------- The research group in Neural Information Processing (Neuroinformatik) at the Universit?t Ulm invites applications for a position of a Research Scientist (Wissenschaftliche/r Mitarbeiter/in) that is open by March 1, 2008. Salary is according to German scale for research staff personnel (TVL-13). The position is initially for 12 months, starting in March 2008 (or soon after), but a full duration of 3 years funding is expected. The position is funded within the EU research project SEARISE. The overall project pursues a highly interdisciplinary approach to develop neural architecture and mechanisms of visual information processing in an observer to analyse motion and form in scenes, group scenic features and deploy attention to salient locations in surveillance scenarios. The announced position here particularly focuses on the development of neural mechanisms of motion integration, how they enter to build attention maps and the development of fast algorithms to demonstrate their real-time performance. An ideal candidate brings in * a Diploma or Master degree in Computer Science, Electrical Engineering, Physics, or related fields, * extensive programming experience (Matlab, C/C++), * sound knowledge in neural models of vision, image processing and/or computer vision, applied mathematics, knowledge about structure and function of the visual cortical architecture, * basic knowledge in pursuing scientific project work, * English speaking and writing skills. The research within the project is expected to be in collaboration with project partners from several European countries and includes on-site visits in different laboratories. It is expected that the candidate pursues a PhD project related to the research topics of the research project. The University of Ulm is an equal opportunity employer. Women are encouraged to apply. Disabled applicants will receive priority in case they have equal qualifications. Please send applications as soon as possible, preferentially via email (PDF document with all the usual data, such as application letter, CV, transcript of records, letters of reference if available) to heiko.neumann at uni-ulm.de. Alternatively, applications can be sent via regular mail to Prof. Heiko Neumann Institut f?r Neuroinformatik Fakult?t f?r Ingenieurwissenschaften und Informatik Universit?t Ulm D-89069 Ulm Germany From friedhelm.schwenker at uni-ulm.de Tue Feb 19 06:59:30 2008 From: friedhelm.schwenker at uni-ulm.de (Friedhelm Schwenker) Date: Tue, 19 Feb 2008 12:59:30 +0100 Subject: Connectionists: PhD Studentship Message-ID: <47BAC4A2.1010303@uni-ulm.de> We invite applications of excellent master students to our Graduate School "Mathematical Analysis of Evolution, Information and Complexity" until March 28, 2008. The detailed announcement can be found here http://www.informatik.uni-ulm.de/ni/stellen/GraduateSchool.pdf Supported Ph.D. work in this call will usually start in May 2008. It may also start at other dates in the periode from April to September 2008. -- Prof. Dr. Guenther Palm Director of the Institute of Neural Information Processing University of Ulm 89069 Ulm Germany phone: ++49 731 50 24150 fax: ++49 731 50 24156 email: guenther.palm at uni-ulm.de From giovanni.pezzulo at istc.cnr.it Wed Feb 20 06:58:38 2008 From: giovanni.pezzulo at istc.cnr.it (Giovanni Pezzulo) Date: Wed, 20 Feb 2008 12:58:38 +0100 Subject: Connectionists: CFP: ABiALS 2008 - The fourth workshop on Anticipatory Behavior in Adaptive Learning Systems Message-ID: Apologies for any multiple copies received. CALL FOR (EXTENDED) ABSTRACTS: ============================================================================ ========== The fourth workshop on Anticipatory Behavior in Adaptive Learning Systems ABiALS 2008 26 June 2008, Munich http://www.psychologie.uni-wuerzburg.de/ABiALS/ (In collaboration with the euCognition Meeting "The Role of Anticipation in Cognition") Deadline for *four-pages abstract* submission: 14 April 2008 Full papers will be published as post-proceedings in the Springer LNAI series ============================================================================ ========== This year, ABiALS invites submission of four-pages (EXTENDED) ABSTRACTS to be discussed (or shown as posters) during the workshop. Abstract submissions should be sent electronically via email in pdf or ps format to: giovanni.pezzulo at istc.cnr.it, indicating in the subject line "ABiALS2008 submission". Submitted abstracts should have a maximal length of FOUR PAGES in 10pt, one-column format. Please use the LNCS Springer-Verlag style at http://www.springeronline.com/comp/lncs/authors.html. Abstracts will be reviewed for acceptance by the program committee and the organizers. After the workshop, *authors of accepted abstracts will be invited to submit a full paper* (up to 20-pages) for post-proceedings publication in the Springer LNAI series, as has been done in 2002 and 2006. Objectives ---------- ABiALS is an interdisciplinary workshop investigating the influence of anticipations on behavior and learning. ABiALS is designed to help investigate how anticipations can influence, initiate, and guide behavior and learning as well as how anticipatory influences can be implemented in an adaptive learning system. In ABiALS 2008, expertise of researchers from various disciplines including neuroscience, cognitive psychology, machine learning, artificial intelligence, control, and vision research are combined to shed further light on the concept of anticipation. Essentially, it is investigated how knowledge about the future influences actual behavior, including influences on attention, action decision making and control, as well as (behavioral and model) learning. Key interests include: ---------------------- -Anticipatory mechanisms and representations for model learning -Model-predictive, adaptive control architectures -Anticipatory, adaptive systems / agents -Distinctions of anticipatory mechanisms -Anticipatory mechanisms in animals and humans List of Topics -------------- -Behavioral psychological and neuroscientific insights on anticipatory behavior and future-oriented capabilities. -Anticipatory functionalities and neural substrates for anticipatory mechanisms in the brain. -Anticipations in decision making and goal-directed behavior. -Trade-off between simple stimulus-response vs. more complex anticipatory driven behavior. -Exploiting anticipations to direct or speed-up motor learning. -Anticipatory mechanisms in motor learning and motor control. -Anticipatory speed-up of sensory processing. -Curiosity and novelty detection during learning and behavior. -Epistemic actions for directed information search and directing attention. -Modular integration of anticipatory mechanisms in cognitive system architectures. -Suitability of predictive representations for anticipatory mechanisms. -Hierarchical predictive and anticipatory architectures. -Anticipations in interaction with motivations and emotions. -Social anticipatory capabilities. -Evolutionary development from reactive to anticipatory capabilities. Important dates --------------- Deadline for abstract submission: 14 April 2008 Notification of acceptance: 14 May 2008 Final version of abstract due: 1 June 2008 ABiALS 2008 Workshop: 26 June 2008 Full paper submission for post-proceedings publication: mid/end September Venue ----- ABiALS 2008 will be held in collaboration with the Fifth Six-Monthly euCognition Meeting "The Role of Anticipation in Cognition" http://www.eucognition.org/six_monthly_meeting_5.htm Venue: Room K13 Municon - Munich Airport Conference Centre (to be confirmed) Important notice ---------------- We are applying for designing ABiALS 2008 as a "euCognition Official Event". In this way euCognition members would be eligible to have all their travel and accommodation costs reimbursed. Applying for euCognition membership is free and very easy. More information will appear soon in the ABiALS website http://www.psychologie.uni-wuerzburg.de/ABiALS/ Program Committee ----------------- We are contacting again the members of the ABiALS 2006 program committee: Christian Balkenius, Edoardo Datteri, Oliver Herbort, Pier Luca Lanzi, Ralf Moeller, Tony Prescott, Jesse Reichler, Alexander Riegler, Deb Roy, Samarth Swarup, Stewart W. Wilson. Organizers ---------- Giovanni Pezzulo Institute of Cognitive Sciences and Technologies (ISTC)-CNR, Via S. Martino della Battaglia 44, I-00185 Rome, Italy. Martin V. Butz Dep. of Psychology, University of W?rzburg, R?ntgenring 11, 97070 W?rzburg, Germany Olivier Sigaud AnimatLab, Laboratoire d'Informatique de Paris 6 (Lip6), 104 avenue de Pr?sident Kennedy, 75016 Paris, France Gianluca Baldassarre Institute of Cognitive Sciences and Technologies (ISTC)-CNR, Via S. Martino della Battaglia 44, I-00185 Rome, Italy. From franco at dii.unisi.it Fri Feb 22 10:21:11 2008 From: franco at dii.unisi.it (Franco Scarselli) Date: Fri, 22 Feb 2008 16:21:11 +0100 Subject: Connectionists: Special Session on "NEURAL NETWORKS IN IMAGE PROCESSING" at KES2008 Message-ID: <47BEE867.3040700@dii.unisi.it> ** Our apologies if you receive multiple copies of this announcement ** Special Session "NEURAL NETWORKS IN IMAGE PROCESSING" 12th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems Zagreb, Croatia 3, 4 and 5 September 2008 Call for papers Neural networks and, more generally, machine learning techniques can be widely applied to image processing and computer vision for preprocessing, feature extraction, segmentation, object tracking and recognition. The adopted machine learning tools include associative memories, kernel machines, neural networks, evolutionary computations, and statistical methods. Visual data are difficult to be dealt with, due to noise, such as light and colour changes, occlusion, and cluttered environments. The capability of machine learning techniques to incorporate imprecision and incomplete information, and to model very complex systems makes them a useful tool in image processing. The aim of this session is to provide an international forum for presentation of recent results and advances regarding soft computing methods in image processing. The scope of this session is, but not limited to: - Noise reduction in images - Image compression - Edge detection and segmentation - Feature extraction - Similarity learning in images - Image analysis and classification - Object localization and detection Paper submission Authors are encouraged to submit high quality, original work that has neither appeared in, nor is under consideration by, other conferences/journals. All submissions will be refereed by experts in the field based on originality, significance, and clarity. The conference preceedings will be published by Springer-Verlag in Lecture Notes in AI as part of the LNCS/LNAI series. When formatting papers - that must be no more than eight (8) pages long, including figures and bibliography - please refer to the Springer-Verlag site and strictly follow the instructions for LNCS authors (http://www.springer.de/comp/lncs/authors.html). Important Dates . Deadline for paper submission: March 10th 2008 . Notification of acceptance: April 10th 2008 . Camera-ready papers: May 1st 2008 Organized by: Chairs . Monica Bianchini, Marco Maggini, Franco Scarselli University of Siena, Italy Program Committee (To be completed) . Annamaria Colla - Elsag Datamat, Italy . Fabio Roli, University of Cagliari, Italy . Carlo Sansone - University of Naples "Federico II", Italy . Lorenzo Sarti, University of Siena, Italy . Joost van de Weijer, GRAVIR-INRIA, France . Ming-Hsuan Yang, Honda Research Institute, USA Other details are available at: http://www.dii.unisi.it/~monica/KES2008_SSNNIP/index.html From franco at dii.unisi.it Fri Feb 22 10:23:25 2008 From: franco at dii.unisi.it (Franco Scarselli) Date: Fri, 22 Feb 2008 16:23:25 +0100 Subject: Connectionists: Special Session on "ADVANCED NEURAL PROCESSING SYSTEMS" at KES2008 Message-ID: <47BEE8ED.4060803@dii.unisi.it> ** Our apologies if you receive multiple copies of this announcement ** Special Session "ADVANCED NEURAL PROCESSING SYSTEMS" 12th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems Zagreb, Croatia 3, 4 and 5 September 2008 Call for papers The research in the neural network field has come to its maturity, both from the theoretical and the practical point of view. Nevertheless, some problems still remain unsolved, since classical connectionist models seem unable to cope with difficult applications, involving complex data. Recent studies on statistical pattern recognition and neural networks show possible directions to exploit structural information in problems which are inherently of topological nature. On the other hand, the actual trend for facing difficult applications is that of using hybrid approaches, integrating in a unique framework neural networks, kernel machines, and statistical techniques. This new scenario imposes studying new models, new learning algorithms, and assessing their properties both theoretically and experimentally. The scope of this session is, but not limited to: - Algorithms and Architectures: statistical learning algorithms, neural networks, kernel methods, dimensionality reduction and manifold learning, model selection. - Applications: innovative applications that use machine learning, including systems for time series prediction, bioinformatics, text/web analysis, multimedia processing, and robotics. - Cognitive Science and Artificial Intelligence: theoretical, computational, or experimental studies of perception, reasoning, problem solving, and natural language processing. - Control and Reinforcement Learning: planning, navigation, Markov decision processes, game-playing. - Learning Theory: generalization, regularization and model selection, new learning paradigms, spaces of functions and kernels, online learning and competitive analysis, hardness of learning and approximations, information theory. - Learning in Graphical Domains: neural network models for graphs, SVMs and kernel methods for graphs, probabilistic models for graphs, statistical relational learning, and pattern recognition applications involving graphical data. Paper submission Authors are encouraged to submit high quality, original work that has neither appeared in, nor is under consideration by, other conferences/journals. All submissions will be refereed by experts in the field based on originality, significance, and clarity. The conference preceedings will be published by Springer-Verlag in Lecture Notes in AI as part of the LNCS/LNAI series. When formatting papers - that must be no more than eight (8) pages long, including figures and bibliography - please refer to the Springer-Verlag site and strictly follow the instructions for LNCS authors (http://www.springer.de/comp/lncs/authors.html). Important Dates . Deadline for paper submission: March 10th 2008 . Notification of acceptance: April 10th 2008 . Camera-ready papers: May 1st 2008 Organized by: Chairs . Monica Bianchini, Marco Maggini, Franco Scarselli University of Siena, Italy Program Committee (To be completed) . Simone Fiori, Politechnic University of Marche, Italy . Barbara Hammer, Clausthal University of Technology, Germany . Alfredo Petrosino, University of Naples "Parthenope", Italy . Friedhelm Schwenker, University of Ulm, Germany . Peter Tino, The University of Birmingham, United Kingdom . Edmondo Trentin, University of Siena, Italy Other details are available at: http://www.dii.unisi.it/~monica/KES2008_SSANPS/index.html From Eugene.Izhikevich at nsi.edu Fri Feb 22 19:45:01 2008 From: Eugene.Izhikevich at nsi.edu (Eugene M. Izhikevich) Date: Fri, 22 Feb 2008 16:45:01 -0800 Subject: Connectionists: Large-scale brain model Message-ID: <47BF6C8D.5090305@nsi.edu> A recent paper describes an ambitious large-scale brain model having multicompartmental spiking neurons, dendritic STDP, 6-layer cortical microcircuitry, and human brain anatomy (using DTI fibertracking). The paper is available at the PNAS site http://www.pnas.org/cgi/content/abstract/0712231105v1 or http://www.izhikevich.com which also shows a movie (158MB) of spiking activity of the model. -- Eugene M. Izhikevich, Senior Fellow in Theoretical Neurobiology PhD, Mathematics, http://www.nsi.edu/users/izhikevich The Neurosciences Institute, Editor-in-Chief at scholarpedia.org 10640 John J. Hopkins Drive tel:(858) 626-2063 San Diego, CA, 92121, USA fax:(858) 626-2099 From s.crone at lancaster.ac.uk Mon Feb 25 12:58:37 2008 From: s.crone at lancaster.ac.uk (Special Conference Session) Date: Mon, 25 Feb 2008 17:58:37 +0000 Subject: Connectionists: Forecasting with Neural Networks Message-ID: <20080225.YOIBHLMJBCKTDBZD@lancaster.ac.uk> ============================================================= FINAL CALL FOR PAPERS - DEADLINE 29 FEBRUARY 2008 ============================================================= FORECASTING WITH NEURAL NETWORKS & COMPUTATIONAL INTELLIGENCE Special Conference Track, Tutorial & Forecasting Competition http://www.neural-forecasting-competition.com/isf_workshop.htm @ the 28th International Symposium on Forecasting June 22-25, 2008, Nice, France http://www.isf2008.org =========================================================== Researchers and Practitioners! This is a final reminder to participate in the conference track on the theory and practice of forecasting with neural networks and computational intelligence at the 2008 International Symposium of Forecasting (ISF'08). ISF'08 will offer you a - Full-day tutorial on forecasting with neural networks (by Hans-Georg Zimmermann, Siemens Research) - Special session track on forecasting with neural networks - Competition on forecasting with neural networks But hurry! The deadline to submit your 300 word abstract is Friday February 29, 2008. ------------------------------------------------------------------- Special Session Topics ------------------------------------------------------------------- We invite abstracts related to the theory and practice of forecasting with neural networks and related methods from computational intelligence, machine learning and data mining, describing new techniques, methods and applications. We strongly encourage submissions from practitioners documenting applications of NN in practice. Particular emphasis is given to submissions related to the use of NN in Information Communication Technology (ICT), the ISF'08 theme. The topics include (but are not limited to): Forecasting Methods - artificial neural networks (all paradigms) - support vector machines & regression - single methods & ensembles - hybrid soft computing methods - etc. Applications - business forecasting & demand planning in supply chains - specific applications, such as electrical load forecasting, call centre forecasting etc. - time series prediction & analysis - predictive classification & data mining - etc. Theory - use of NN in changing environments - stability & retraining of networks - principles of modelling NN for forecasting - data-preprocessing for NN forecasting - model selection of NN candidates - etc. ------------------------------------------------------------------- Forecasting Competition CfP ------------------------------------------------------------------- In addition to the general track, we invite submissions to the NN5 time series forecasting competition (www.neural-forecasting-competition.com) open to neural networks and all methods of computational intelligence. The 2008 NN5 competition focuses on predicting 11 or 111 daily time series of cash money withdrawals at cash machines in England. Submissions to the NN5 competition will also be presented in special sessions at the ISF'08 conference. To present results each participant must submit an abstract to ISF'08, register for the conference and submit their predictions at the competition website. More information and datasets may be found at the NN5 competition ISF'08 forecasting workshop webpage (http://www.neural-forecasting-competition.com/isf_workshop.htm). ------------------------------------------------------------------- Submission of Abstracts / Papers ------------------------------------------------------------------- Prospective authors submitting to the NN5 forecasting competition as well as the neural network forecasting special sessions are invited to submit their abstract (limit of 300 words) electronically on the conference website http://forecasters.org/isf/abstracts.html following the guidelines posted there. For the forecasting competition, the abstract should describe the NN model and modelling process applied and contain the reference "NN5 forecasting competition" in the abstract title. Please submit all abstracts using the "Subject Area" = "Neural Nets". All abstract submissions must be received by February 29th, 2008. Full submissions of papers will be encouraged after the conference. ------------------------------------------------------------------- Conference Information ------------------------------------------------------------------- The 28th International Symposium on Forecasting (ISF'08) (at www.forecasters.org/isf) - organised annually by the International Institute of Forecasting (www.forecasters.org) and now running in its 28th successful year - is the premier multidisciplinary forecasting conference, attracting the world's leading forecasting researchers, practitioners, and students. Through a combination of keynote speaker presentations, academic sessions, workshops, and social programs, the ISF provides many excellent opportunities for networking, learning, and fun. Past conferences have attracted well over 200 presentations in all areas of forecasting, often with more then 30 presentations dedicated to neural networks for forecasting in various sessions on NN theory, NN applications and a NN forecasting competition. It comprises one of the largest gatherings focussed on neural forecasting and provides a premier opportunity for networking within the field. All recent ISF conferences have hosted a dedicated track on NN in addition to various submissions to alternative tracks of nonlinear methods, and application focused tracks such as financial forecasting, forecasting in sports etc. The 28th Annual International Symposium on Forecasting will be held June 22-25, 2007 at the Nice SAS Radisson Plaza. Papers on all aspects of forecasting are welcome. The objectives of the symposium are to help disseminate knowledge about new developments in statistical and judgmental forecasting methods and to promote discussion about forecasting among academicians and practitioners. The organizers welcome the attendance of delegates from diverse fields, including Economics, Finance, Business, Public Policy, Statistics, Climate and Environment, Information Systems, Psychology, Demography, Political Science, Energy and Computational Intelligence. ------------------------------------------------------------------- Important Dates ------------------------------------------------------------------- February 29, 2008 Abstracts due March 28, 2008 Notification of acceptance April 18, 2008 Early registration deadline (for discounts) June 22-25, 2008 The International Symposium on Forecasting 2008 August 2008 Call for Post-conference publications We look forward to welcoming you in Nice! Sven F. Crone ______________________________________________________ Sven F. Crone Deputy Director, Lancaster Centre for Forecasting Assistant Professor in Management Science (Lecturer) Lancaster University Management School Department of Management Science Lancaster LA1 4YX, United Kingdom Tel +44 (0)1524 592991direct Tel +44 (0)1524 593867 department - Fax +44 (0)1524 844885 Internet http://www.lums.lancs.ac.uk - eMail s.crone at neural-forecasting.com _______________________________________________ From YOMTOV at il.ibm.com Thu Feb 28 11:15:55 2008 From: YOMTOV at il.ibm.com (Elad Yom-Tov) Date: Thu, 28 Feb 2008 18:15:55 +0200 Subject: Connectionists: Pascal Large Scale Learning Challenge Message-ID: Apologies for any cross-postings. Pascal Large Scale Learning Challenge ================================== The Large Scale Learning challenge is aimed at comparing the scalability and efficiency of existing classification algorithms when applied to large data. The goal of this challenge is to encourage research in Machine Learning and related applications by providing several large datasets, uniform scoring procedures, and a forum for organizations interested in comparing their results. The experience in fields such as Information Retrieval has shown that such challenges are extremely useful in advancing the state of the art and in fostering new approaches to the problem. Time Line ========= 27 February 2008 Challenge Announcement 12 June 2008 Test sets available 26 June 2008 End of the competition 27 June - 8 July Re-calibration of all entries 9 July 2008 ICML'08 Large Scale Learning workshop 10 July - ... Re-run of the top ten methods Call for Participation ====================== We encourage participants to download the datasets, submit their results to the challenge web site for instant feedback on their performance over a validation set, and to send their predictions for classification of a test set. Prizes ====== We will award travel grants or cash prizes to the winners. Evaluation ========== The results of the challenge will be discussed at an ICML'08 workshop, where the best and most innovative approaches will be presented. Participants will be invited to submit to submit a paper to a JMLR special topic on large scale learning (*TBC*). Detailed information can be found at: http://largescale.first.fraunhofer.de/