From njm@cupido.inesc.pt Mon Sep 18 14:52:12 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Mon, 18 Sep 95 14:51:31 -0500; AA00238 Received: from bilby.cs.uwa.oz.au by lucy.cs.wisc.edu; Mon, 18 Sep 95 14:47:18 -0500 Received: from (mafm@parma.cs.uwa.oz.au [130.95.1.7]) by cs.uwa.oz.au (8.6.8/8.5) with SMTP id TAA02232; Mon, 18 Sep 1995 19:55:08 +0800 Message-Id: <199509181155.TAA02232@cs.uwa.oz.au> From: Nuno Joao Mamede To: reinforce@cs.uwa.edu.au Subject: FINAL PROGRAM: EPIA'95 Date: Sun, 17 Sep 95 20:27:59 +0100 EPIA'95 - PRELIMINARY PROGRAM SEVENTH PORTUGUESE CONFERENCE ON ARTIFICIAL INTELLIGENCE Casino Park Hotel, Funchal, Madeira Island, Portugal 3-6 October, 1995 (Under the auspices of the Portuguese Association for AI) The 7th Portuguese Conference on Artificial Intelligence will be held at Funchal, Madeira Island, Portugal, October 3-6, 1995. As in the past, EPIA 95 is an international conference with English as the official language. The conference covers all areas of Artificial Intelligence, including theoretical areas, foundational areas, and applications. The scientific program consists of invited lectures, tutorials, demonstrations and paper presentations. There will also be parallel workshops on Expert Systems, Fuzzy Logic and Neural Networks, and Applications of AI to Robotics and Vision Systems. ====================================================================== FINAL PROGRAM (Summary) ====================================================================== Tuesday - October, 3 95 ~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 12:30 TUTORIAL 1 - Artificial Life and Autonomous Robots Luc Steels 9:00 - 12:30 TUTORIAL 3 - Introduction to Artificial Intelligence Ernesto Costa (in Portuguese) 14:30 - 18:00 TUTORIAL 2 - Virtual Reality - The AI perspective David Hogg 14:30 - 18:00 TUTORIAL 4 - Design of Expert Systems Ernesto Morgado (in Portuguese) Wednsday - October, 4 95 ~~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 9:30 OPENING SESSION 9:40 - 10:30 QUALITATIVE REASONING 10:30 - 10:50 Coffee break 10:50 - 12:30 NEURAL NETWORKS & DISTRIBUTED ARTIFICIAL INTELLIGENCE 10:50 - 12:30 FUZZY LOGIC & NEURAL NETWORKS WORKSHOP 12:30 - 14:00 Lunch 14:00 - 15:30 Invited Lecture by LUIS B. ALMEIDA (IST - Portugal) "The Connectionist Paradigm and AI" 15:30 - 15:50 Coffee break 15:50 - 18:00 BELIEF REVISION & NON-MONOTONIC REASONING 15:50 - 18:30 FUZZY LOGIC & NEURAL NETWORKS WORKSHOP 15:50 - 18:30 APPLICATIONS OF EXPERT SYSTEMS WORKSHOP 20:00 - Welcome Dinner (With a performance of the Univ. of Madeira "tuna") Thursday - October, 5 95 ~~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 10:30 Invited Lecture by RODNEY BROOKS (MIT - USA) "The Evolutionist Approach - Past, Present and Future of AI" 10:30 - 10:50 Coffee break 10:50 - 12:30 APPLICATIONS OF EXPERT SYSTEMS WORKSHOP 10:50 - 11:40 ROBOTICS AND CONTROL 11:40 - 12:30 POSTER SECTION 12:30 - 14:00 Lunch 14:00 - 15:15 MACHINE LEARNING 15:15 - 15:35 Coffee break 15:35 - 17:05 Invited Lecture by MARVIN MINSKY (MIT - USA) "Why Human Brains Can't Really Think" 17:15 - 18:30 Visit to the Madeira Wine Cellars Friday - October, 6 95 ~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 10:30 INVITED LECTURE by Manuela Veloso (CMU - USA) "Planning and Learning in Intelligent Agents" 10:30 - 10:50 Coffee break 10:50 - 12:30 STREAM 1: PLANNING AND CASE-BASED REASONING 10:50 - 12:30 STREAM 2: CONSTRAINT-BASED REASONING 10:50 - 12:30 APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP 12:30 - 14:00 Lunch 14:00 - 15:30 STREAM 1: AUTOMATED REASONING AND THEOREM PROVING 14:00 - 15:30 STREAM 2: GENETIC ALGORITHMS & THEORY OF COMPUTATION 14:00 - 15:30 APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP 15:30 - 15:50 Coffee break 15:50 - 17:30 PANNEL (The Next Frontiers of AI: the Role of Foundations) 18:00 - 19:00 APPIA meeting 20:00 Farewell Dinner (with folklore dances show) Saturday - October, 7 95 ~~~~~~~~~~~~~~~~~~~~~~~~ TOUR 1 - Island Tour (full day) TOUR 2 - Ribeiro Frio/Portela Walking Tour (full day) TOUR 3 - Eira do Serrado (half day) ====================================================================== PAPERS TO BE PRESENTED IN EACH SESSION ====================================================================== AUTOMATED REASONING AND THEOREM PROVING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Terminological Meta-Reasoning by Reification and Multiple Contexts Klemens Schnattinger, Udo Hahn, Manfred Klenner CLIF, Freiburg University, Germany A New Continuous Propositional Logic Riccardo Poli, Mark Ryan, Aaron Sloman SCS, The University of Birmingham, UK Super-Polynomial Speed-Ups in Proof Length by New Tautologies Uwe Egly FG Intellektik, TH Darmstadt, Germany BELIEF REVISION ~~~~~~~~~~~~~~~ Belief Revision in Non-Monotonic Reasoning Jose Alferes, Luis Moniz Pereira, T. Przymusinski DM, U. Evora, and CRIA, U. Nova de Lisboa, Portugal, and University of California at Riverside, USA A New Representation of JTMS Truong Quoc Dung IRIDIA, Universite Libre de Bruxelles, Belgium CONSTRAINT-BASED REASONING ~~~~~~~~~~~~~~~~~~~~~~~~~~ The Retrieval Problem in a Concept Language with Number Restrictions Aida Vitoria, Margarida Mamede, Luis Monteiro DI, Universidade Nova de Lisboa, Portugal Formalizing Local Propagation in Constraint Maintenance Systems Gilles Trombettoni INRIA-CERMICS, France A Dependency Parser of Korean Based on Connectionist/Symbolic Techniques Jong-Hyeok Lee, Geunbae Lee Pohang University of Science and Technology, Korea A Symbiotic Approach to Arc and Path Consistency Checking Pierre Berlandier INRIA-CERMICS, France DISTRIBUTED ARTIFICIAL INTELLIGENCE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Where Do Intentions Come From?: A Framework for Goals and Intentions Adoption, Derivation and Evolution Graca Gaspar, Helder Coelho Faculdade de Ciencias de Lisboa, and INESC, Portugal A Closer Look to Artificial Learning Environments Helder Coelho, Augusto Eusebio, Ernesto Costa INESC, Portugal, and DEI, Universidade de Coimbra, Portugal Building Multi-Agent Societies from Descriptions to Systems: Inter-Layer Translations Helder Coelho, Luis Antunes, Luis Moniz INESC, Portugal GENETIC ALGORITHMS ~~~~~~~~~~~~~~~~~~ GA/TS: A Hybrid Approach for Job Shop Scheduling in a Production System Jose Ramon Zubizarreta, Javier Arrieta Facultad de Informatica de San Sebastian, Spain MACHINE LEARNING ~~~~~~~~~~~~~~~~ A Controlled Experiment: Evolution for Learning Difficult Image Classification Astro Teller, Manuela Veloso Carnegie Mellon University, USA Minimal Model Complexity Search Chris McConnell CMU School of Computer Science, USA Characterization of Classification Algorithms Joao Gama, Pavel Brazdil LIACC, Universidade do Porto, Portugal NEURAL NETWORKS ~~~~~~~~~~~~~~~ Neurons, Glia and the Borderline Between Subsymbolic and Symbolic Processing J. G. Wallace, K. Bluff Swinburne University of Technology, Australia NON-MONOTONIC REASONING ~~~~~~~~~~~~~~~~~~~~~~~ Arguments and Defeat in Argument-Based Nonmonotonic Reasoning Bart Verheij University of Limburg, The Netherlands A Preference Semantics for Ground Nonmonotonic Modal Logics Daniele Nardi, Riccardo Rosati DIS, Universita di Roma ``la Sapienza", Italy Logical Omniscience vs. Logical Ignorance On a Dilemma of Epistemic Logic Ho Ngoc Duc ILPS, University of Leipzig, Germany PLANNING AND CASE-BASED REASONING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ On the Role of Splitting and Merging Past Cases for Generation of New Solutions Carlos Bento, Penousal Machado, Ernesto Costa DEI, Universidade de Coimbra, Portugal Theorem Proving by Analogy - A Compelling Example Erica Melis Department of AI, University of Edimburgh, Scotland Non-Atomic Actions in the Situation Calculus Jose Julio Alferes, Renwei Li, Luis Moniz Pereira CRIA and DCS, Universidade Nova de Lisboa, Portugal Planning under Uncertainty: A Qualitative Approach Nikos Karacapilidis FIT.KI, GMD, Sankt Augustin, Germany QUALITATIVE REASONING ~~~~~~~~~~~~~~~~~~~~~ Qualitative Reasoning under Uncertainty Daniel Pacholczyk DMI, U.F.R., Science d'Angers, France Systematic Construction of Qualitative Physics-Based Rules for Process Diagnostics Jaques Reifman, Thomas Y.C. Wei Argonne National Laboratory, USA ROBOTICS AND CONTROL ~~~~~~~~~~~~~~~~~~~~ Integrated Process Supervision (IPS): A Structured Approach to Expert Control Chai Quek, P.W. Ng, M. Pasquier Nanyang Technological University, Singapore Using Stochastic Grammars to Learn Robotic Tasks Pedro Lima, George Saridis ISR, Technical Univ. of Lisbon, Portugal, and Rensselaer Polytechnic Institute, USA THEORY OF COMPUTATION ~~~~~~~~~~~~~~~~~~~~~ Constraint Categorial Grammars Luis Damas, Nelma Moreira LIACC, Universidade do Porto, Portugal A New Translation Algorithm from Lambda Calculus into Combinatory Logic Sabine Broda, Luis Damas LIACC, Universidade do Porto, Portugal POSTER SECTION ~~~~~~~~~~~~~~ Interlocking Multi-Agent and Blackboard Architectures Bernhard Kipper DCS, University of Saarbrucken, Germany A Model Theory for Paraconsistent Logic Programming Carlos Viegas Damasio, Luis Moniz Pereira CRIA, and DCS, Universidade Nova de Lisboa, Portugal Promoting Software Reuse Through Explicit Knowledge Representation Carmen Fernandez-Chamizo, Pedro A. Gonzalez-Calero, Mercedes Gomez-Albarran Universidad Complutense, Spain Efficient Learning in Multi-Layered Perceptron Using the Grow-And-Learn Algorithm Gildas Cherruel, Bassel Solaiman, Yvon Autret Univ. de Bretagne Occidentale, and TNI, and ENSTB, France An Non-Diffident Combinatorial Optimization Algorithm Gilles Trombettoni, Bertrand Neveu INRIA-CERMICS, France Modelling Diagnosis Systems with Logic Programming Iara Mora, Jose Alferes CRIA, U. Nova de Lisboa, and DM, U. Evora, Portugal Agreement: A Logical Approach to Approximate Reasoning Luis Custodio, Carlos Pinto-Ferreira ISR, Technical University of Lisbon, Portugal Constructing Extensions by Resolving a System of Linear Equations Messaoudi Nadia Universite Aix-Marseille II, France Presenting Significant Information in Expert System Explanation Michael Wolverton Daresbury Rutherford Appleton Laboratory, UK A Cognitive Model of Problem Solving with Incomplete Information Nathalie Chaignaud LIPN, Universite Paris-Nord, France Filtering Software Specifications Written In Natural Language Nuria Castell, Angels Hernandez Universitat Politecnica de Catalunya, Spain Parsimonious Diagnosis in SNePS Pedro A. Matos, Joao P. Martins DEM, Technical University of Lisbon, Portugal Syntactic and Semantic Filtering in a Chart Parser Sayan Bhattacharyya, Steven L. Lytinen University of Michigan, and DePaul University, USA GA Approach to solving Multiple Vehicle Routing Problem Slavko Krajcar, Davor Skrlec, Branko Pribicevic, Snjezana Blagajac Faculty of Electrical Eng. and Computing, Croatia Multilevel Refinement Planning in an Interval-Based Temporal Logic Werner Stephan and Susanne Biundo German Research Center for AI, Germany APPLICATIONS OF EXPERT SYSTEMS WORKSHOP ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Wednesday, 4 15:50 - 16:15 CGD - An Expert System for Loan Analysis Decision Vasco Moreira, Andre Frazao, Elisabete Silva, Ernesto Costa Caixa Geral de Depositos, Dir. Organizacao Informatica Lisboa, Portugal 16:15 - 16:40 A Quantitative Method for Performing A Cost-Benefit Analysis of Expert System Projects Ramu Kannan, Reza Khorramshahgol, Mohan Tanniru Dept. of Management Science and Economics, Coppin State College, Baltimore, USA 16:40 - 17:05 DARE: a Knowledge-Based System for the Diagnosis of Neuromuscular Disorders J. Cruz, P. Barahona, A. P. Figueiredo, M. Veloso, M. Carvalho UNINOVA, Portugal 17:05 - 17:30 A Cooperative Multi-Agent System for Strategic Decision Making Suzanne Pinson Jorge Louca Universite Paris IX - Dauphine, Paris, France 17:30 - 17:55 A Hybrid Model for Classification Expert Systems Sergio Rosa, Beatriz Leao Instituto de Informatica UFRGS, Porto Alegre, Brasil Thursday, 5 10:50 - 11:15 PERMEX - Expert System for Corrosion Failure Analysis Fernando Lopes, A. Novais, N. Mamede, C. Rangel INETI, DMS, Lisboa, Portugal 11:15 - 11:40 Architectural Aspects of an Intelligent DSS for Flow Shop Production Control Ioannis Hatzilygeroudis, D. Sofotassios, N. Dendris, P. Spirakis, A. Tsakalidis Dept. of Computer Engin. and Informatics, Univ. of Patras, Greece 11:40 - 12:05 Advances in Explanation Facilities for Expert Systems Keith Darlington School of Computing, Information Systems and Mathematics South Bank University, London, UK 12:05 - 12:30 A Deferred Communication in a Parallel Distributed Expert System Shell Wided Lejouad SECOIA Project, Sophia Antipolis, France APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Friday, 6 10:50 - 11:15 Designing and Implementing Real Walking Agents Using Virtual Environments Aleix Martinez Universitat Autonoma de Barcelona, Dept. Informatica, Spain 11:15 - 11:40 Multi-Layer Perceptrons for Task Visual Servoing in Robotics Nadine Rondel, Gilles Burel Thomson CSF-LER, France 11:40 - 12:05 Heuristic Autonomous Mobile Robot Using Visual Servoing Jean-Charles Bonin, Fernandoo De Carvalho Gomes Laboratorio de Inteligencia Artificial-LIA, Fortaleza, Brasil 12:05 - 12:30 An Integrated Approach to Position a Robot Arm in a System for Planar Part Grasping Pedro Sanz, Juan Domingo Universitat Jaume I, Dpto. Informatica, Castellon, Spain 14:00 - 14:25 Learning and Recall of Robot Manipulator Motions Using Driver Programs Frank Smieja, Uwe Bayer GMD, Schloss Birlinghoven, Germany 14:25 - 14:50 Selective Visual Perception Driven by Cues from Speech Processing Reinhard Moratz AG Angewandte Informatikj, Universitaet Bielefeld, Germany 14:50 - 15:15 Autonomous Robots and Active Vision Systems: Issues on Architectures an Integration Helder Araujo, Jorge Dias, Jorge Batista, Paulo Peixoto ISR-Coimbra, Universidade de Coimbra, Portugal 15:15 - 15:35 Learning from Perception, Success and Failure in a Team of Autonomous Mobile Robots Arvin Agah, George Bekey Inst. for Robotics and Intell.Syst., Univ. of Southern California, USA FUZZY LOGIC & NEURAL NETWORKS WORKSHOP IN ENGINEERING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Wednesday, 4 - Session 1 10:50 - 11:15 A Fuzzy Logic Controller for Supraconductivity Measuring N. Zimic, J. Ficzko, M. Mraz, J. Virant Faculty of Electrical & Computer Engineering Science University of Ljubljana, Slovenia 11:15 - 11:40 Complex Data and Fuzziness in Database Applications Adnan Yazici Dept. of Computer Engineering, Middle East Technical University, Turkey 11:40 - 12:05 Car License Plate Recognition with Neural Networks and Fuzzy Logic J. Nijhuis, et.al. Dept. of Computer Science, Groningen University, The Netherlands 12:05 - 12:30 Similarity-Based Self-organized Clustering Jurgen Rahmel Center for Learning Systems & Applications, University of Kaiserslautern, Germany 15:50 - 16:15 On the Representation of Data for Optimal Learning M. Brugge, J. Nijhuis, W. Jansen, H. Drenth, L. Spaanenburg Dept. of Computer Science, Groningen University, The Netherlands 16:15 - 16:40 Artificial Neural Net-Based Controllers for Real Process Control Petr Pivonka, Jan Zizka Dept. of Automatic Control and Instrumentation Technical University of Brno, Czech Republic 16:40 - 17:05 A Production Line for Generating Clinical Decision Support Systems Patrik Eklund Dept. of Computing Science, Umea University, Sweden 17:05 - 17:30 Knowledge Discovery Using Hierarchical Connectionist Marie Pai, Robin Ying AT&T Bell Laboratories, USA 17:30 - 17:55 Growing Filters for Finite Impulse Response Networks M. Diepenhorst, J. Nijhuis, R. Venema, L. Spaanenburg Dept. of Computer Science, Groningen University, The Netherlands ====================================================================== ENQUIRIES ADDRESS ====================================================================== EPIA'95 - INESC E-mail: epia95@inesc.pt Av. Alves Redol, 9 Fax: 351-1-525843 1000 Lisboa Voice: 351-1-3100325 PORTUGAL Home Page: http://www.isr.ist.utl.pt/~cpf/epia95 ====================================================================== SUPPORTERS ====================================================================== Banco Nacional Ultramarino Governo Regional da Madeira Instituto Superior Tecnico SISCOG - Sistemas Cognitivos INESC CITMA IBM TAPair Portugal =========================== Last Line ================================ From pfbaldi@cco.caltech.edu Tue Sep 19 15:55:56 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Tue, 19 Sep 95 15:55:37 -0500; AA16373 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Tue, 19 Sep 95 15:55:33 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa25416; 19 Sep 95 14:18:46 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa25414; 19 Sep 95 13:50:59 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa04595; 19 Sep 95 13:50:24 EDT Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id aa10826; 19 Sep 95 11:35:09 EDT Received: from chamber.cco.caltech.edu by EDRC.CMU.EDU id aa23875; 19 Sep 95 11:33:58 EDT Received: from accord.cco.caltech.edu by chamber.cco.caltech.edu with ESMTP (8.6.12/DEI:4.41) id IAA10169; Tue, 19 Sep 1995 08:33:30 -0700 Received: by accord.cco.caltech.edu (8.6.7/UGCS:4.41) id IAA08247; Tue, 19 Sep 1995 08:33:27 -0700 Date: Tue, 19 Sep 1995 08:33:26 -0700 (PDT) From: Pierre Baldi To: Connectionists@cs.cmu.edu Cc: mozer@neuron.cs.colorado.edu, asl@t13.lanl.gov, ramit@t10.lanl.gov, pfbaldi@accord.cco.caltech.edu Subject: Tal Grossman Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Tal Grossman tragically died in a car accident on August 1st. Tal was in the Complex Systems Group at Los Alamos. He was very active in the area of machine learning and computational molecular biology. He gave a presentation at one of the NIPS workshops last year. This year he was trying to organize the same workshop himself. He is survived by his wife and children. Anyone who knew Tal, and feels the need to, is welcome to contact either his wife: Dr. Ramit Mehr-Grossman ramit@t10.lanl.gov or his sponsor at Los Alamos: Dr. Alan Lapedes asl@t13.lanl.gov Pierre Baldi From igor@c3serve.c3.lanl.gov Wed Sep 20 01:37:00 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Wed, 20 Sep 95 01:36:58 -0500; AA21664 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Wed, 20 Sep 95 01:36:57 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa25797; 19 Sep 95 21:04:46 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa25795; 19 Sep 95 20:52:44 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa04847; 19 Sep 95 20:51:55 EDT Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id aa17853; 19 Sep 95 19:49:04 EDT Received: from c3serve-e2.c3.lanl.gov by EDRC.CMU.EDU id aa26022; 19 Sep 95 19:48:36 EDT Received: from c3serve.c3.lanl.gov (monsoon.c3.lanl.gov [128.165.21.52]) by c3serve.c3.lanl.gov (8.6.10/93112601) with ESMTP id RAA16530 for ; Tue, 19 Sep 1995 17:48:30 -0600 Message-Id: <199509192348.RAA16530@c3serve.c3.lanl.gov> To: connectionists@cs.cmu.edu Subject: postdoctoral positions available Date: Tue, 19 Sep 1995 17:48:30 -0600 From: Igor Zlokarnik POSTDOCTORAL POSITIONS IN STATISTICAL ANALYSIS AVAILABLE Los Alamos National Laboratory Los Alamos, NM 87545 Postdoctoral positions are available to participate in the development of appropriate methods for analysing large databases such as medical payment records and related databases for the purpose of detecting and preventing fraud, waste and abuse. This research includes, but is not limited to, the evaluation and modification of existing methods, such as neural networks, genetic algorithms, fuzzy logic, n-grams, multivariate analysis, factor analysis, and multidimensional scaling. It may also involve the implementation of database interfaces. Los Alamos National Laboratory provides excellent opportunities for advanced research. The Laboratory operates the world's largest scientific computing facility. A major strength of Los Alamos is the interdisciplinary nature of much of it research. Scientists in one field may draw on research and techniques developed for quite a different, seemingly unrelated, area. Your immediate team colleagues will be working on such diverse research areas as automatic speech recognition, virtual reality, financial analysis, etc. Appointments are available for applicants who have received a doctoral degree in the past three years or will have completed all PhD requirements by date of hire. Positions are for 2 years and are renewable for a third year. Salaries range between 40k - 45k per annum depending on the number of years since the PhD was earned. Los Alamos National Laboratory is an equal opportunity/affirmative action employer. It is operated for the Department of Energy by the University of California. Applicants must submit a resume including list of publications, a statement of research interests, and three letters of recommendation to: Dr. George Papcun Los Alamos National Laboratory CIC-3, MS B256 Los Alamos, NM 87545 phone: (505) 667-9800 e-mail: gjp@lanl.gov by no later than 15 November 1995. Preliminary E-mail inquiries are encouraged. From asl@santafe.edu Wed Sep 20 17:39:23 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Wed, 20 Sep 95 17:39:21 -0500; AA04646 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Wed, 20 Sep 95 17:39:17 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa25931; 20 Sep 95 0:22:32 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa25929; 20 Sep 95 0:10:56 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa04944; 20 Sep 95 0:10:13 EDT Received: from CS.CMU.EDU by B.GP.CS.CMU.EDU id aa20776; 20 Sep 95 0:07:55 EDT Received: from sfi.santafe.edu by CS.CMU.EDU id aa20927; 20 Sep 95 0:07:34 EDT Received: from santafe (santafe.santafe.edu) by sfi.santafe.edu (4.1/SMI-4.1) id AA18984; Tue, 19 Sep 95 22:06:14 MDT Date: Tue, 19 Sep 95 22:06:14 MDT From: Alan Lapedes Message-Id: <9509200406.AA18984@sfi.santafe.edu> To: Connectionists@cs.cmu.edu Subject: Tal Grossman It is with deepest regret that we have to announce that Tal Grossman, a postdoctoral fellow in the Complex Systems Group at Los Alamos, was killed August 1, 1995 in a car accident while on a family vacation in Arizona. His wife, Ramit (a postdoctoral fellow in the Theoretical Biology Group at Los Alamos) and the children have recovered from minor injuries and are all right. Tal was very active in neural networks and computational biology and had a very promising career. The funeral was held in Israel. A memorial service will be held in Los Alamos Wed Sept 20, 1995. If they wish, friends and colleagues of the Grossmans may contact either Ramit, or Alan Lapedes (Tal's postdoctoral superviror) at: ramit@t10.lanl.gov asl@t13.lanl.gov From cia@kamo.riken.go.jp Wed Sep 20 17:39:33 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Wed, 20 Sep 95 17:39:26 -0500; AA04652 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Wed, 20 Sep 95 17:39:21 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa26887; 20 Sep 95 13:49:12 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa26885; 20 Sep 95 13:34:30 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa05649; 20 Sep 95 13:33:59 EDT Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id aa28250; 20 Sep 95 10:36:44 EDT Received: from kamo.riken.go.jp by EDRC.CMU.EDU id aa28835; 20 Sep 95 10:35:50 EDT Received: by kamo.riken.go.jp; id AA15387; Wed, 20 Sep 1995 23:40:23 +0900 Message-Id: <9509201440.AA15387@kamo.riken.go.jp> To: connectionists@cs.cmu.edu Cc: cia@zoo.riken.go.jp Subject: Blind Separation of source (abstracts) Date: Wed, 20 Sep 95 23:40:22 +0900 From: cia@kamo.riken.go.jp X-Mts: smtp Blind Signal Processing is an emerging area in adaptive signal processing and neural networks. It was originated in France in the late 80's . Below please find an advanced program of a special invited session devoted to blind separation of sources and their applications at 1995 INTERNATIONAL SYMPOSIUM ON NONLINEAR THEORY AND ITS APPLICATIONS , NOLTA'95 in Las Vegas. Any comments will be highly appreciated, especially association of this approach to brain information processing and image and speech enhancement, filtering and noise reduction. Andrzej Cichocki, Head of Laboratory for Artificial Brain Systems, Frontier Research Program RIKEN, Institute of Physical and Chemical Research, Hirosawa 2-1, Saitama 351-01, WAKO-Schi, JAPAN E-mail: cia@kamo.riken.go.jp, FAX (+81) 048 462 4633. URL: http://zoo.riken.go.jp/bip.html =========================================================================== NOLTA'95, 1995 INTERNATIONAL SYMPOSIUM ON NONLINEAR THEORY AND ITS APPLICATIONS Caesars Palace, LAS VEGAS Dec. 10 -14, 1995 Program for Special Invited Session on "BLIND SEPARATION OF SOURCES -Brain Information Processing" Organizer and chair Dr. A. Cichocki Frontier Research Program RIKEN, Institute of Physical and Chemical Research, Hirosawa 2-1, Saitama 351-01, WAKO-Schi, JAPAN Advanced Program: 1. Prof.Christian JUTTEN , Laboratory TIRF, INPG, Grenoble, FRANCE, "Separation of Sources: Blind or Unsupervised? " Abstract: Basically, separation of sources are referred as BLIND methods. However, adaptive algorithms for source separation only emphasize on UNSUPERVISED aspects of the learning. In this talk, we propose a selected review of recent works to show how A PRIORI KNOWLEDGES on the sources or on the mixtures can simplify the algorithms and improve performance. 2. Prof. Jean-Francois CARDOSO , Ecole Nationale Superieure des Telecommunications, Telecom Paris, FRANCE "The Invariant Approach to Source Separation" Abstract: The `invariant approach' to source separation is based on the recognition that the unknown parameter in a source mixture is the mixing matrix, hence it belongs to a multiplicative group. In this contribution, we show that this simple fact can be exploited to build source separation algorithms behaving uniformly well in the mixing matrix. This is achieved if two sufficient conditions are met. + First, contrast functions (or estimating equations) used to identify the mixture should be designed in such a way that source separation is achieved when they are optimized (or solved) **without constraints** (such as normalization, etc). Examples of such contrast functions will be given, some of them being simple variants of classic contrast functions. This requirement is sufficient to guarantee uniform performance of the resulting batch algorithms. + Second, in the case of adaptive algorithm, uniform performance has a more extensive meaning: not only the residual error but also the convergence are important. Again, the multiplicative nature of the parameter calls for a special form of the learning rule, namely it suggests a `multiplicative update'. This approach results in adaptive source separation algorithms and enjoying uniform performance: convergence speed, residual error, stable points, etc... do not depend on the mixing matrix. In addition these algorithms show a very simple (and parallelizable) structure. The paper includes analytical results based on asymptotic performance analysis that quantify the behavior of both batch and adaptive equivariant source separators. In particular, these results allow to determine, given the source distribution, the optimal nonlinearities to be used in the learning rule. 3. Dr. Jie ZHU, Prof. Xi-Ren CAO, and prof. Ruey-Wen LIU, The Hong Kong University of Science and Technology Kowloon, Hong Kong, The University of Notre Dame, Notre Dame, NI 46556, U.S.A. "Blind Source Separation Based on Output Independence - Theory and Implementation" Abstract: The paper presents some recent results on the theory and implementation techniques of blind source separation. the approach is based on independence property of the outputs of a filter. In the theory part, we identify and study two major issues in the blind source separation problem: separability and separation principles. We show that separability is an intrinsic property of the measured signals and can be described by the concept of $m$-row decomposability introduced in this paper, and that the separation principles can be developed by using the structure characterization theory of random variables. In particular, we show that these principles can be derived concisely and intuitively by applying the Darmois-Skitovich theorem, which is well-known in statistical inference theory and psychology. In the implementation part, we show that if at most one of the source signals has a zero third (or fourth) order cumulant, then these signals can be separated by a filter whose parameters can be determined by a system of nonlinear equations using only third (or fourth) order cumulants of the measured signals. This results covers some previous results as special cases. 4. Dr. Jie HUANG, Prof. Noboru OHNISHI and Dr. Noboru Sugie ; Bio-Mimetic Control Research Center , The Institute of Physical and Chemical Research (RIKEN), Nagoya, JAPAN "Sound Separation Based on Perceptual Grouping of Sound Segments" Abstract : We would like to propose a sound separation method, which combines spatial cues (source direction) and structural cues (continuity and harmony). Sound separation is important in various scientific fields. There are mainly two different approaches to achieve this goal. One is based on blind estimation of inverse transfer functions from multiple sources to multiple receivers (microphones). The other is based on grouping sound segments in time-frequency domain. Our approach is based on the sound segments grouping. However, we use multiple microphones to obtain the spatial information. This approach is strongly inspired by the precedence effect and the cocktail party effect of human auditory system. The precedence effect suggests to us the way of coping with echoes in reverberant environment. The cocktail party effect suggests the use of spatial cues for sound separation. Psychological factors of auditory stream integration and segregation, such as continuity and harmony, are used as structural cues. It is realized by a continuity enhancement filter and a harmonic histogram to supplement the spatial segments grouping. The use of this method with real human speeches recorded in an anechoic chamber and a normal room was demonstrated. The experiments have shown that the method was effective to separate sounds in reverberant environments. 5. Dr. Kiyotoshi MATSUOKA and Dr. Mitsuru KAWAMOTO, Department of Control Engineering, Kyushu Institute of Technology, 1-1,Tobata, Kitakyushu, 804 Japan "Blind Signal Separation Based on a Mutual Information Criterion" Abstract: This paper deals with the problem of the so-called blind separation of sources. The problem is to recover a set of source signals from their linear mixtures observed by the same number of sensors, in the absence of any particular information about the transfer function that couples the sources and the sensors. The only a priori knowledge is, basically, the fact that the source signals are statistically mutually independent. Such a task arises in noise canceling of sound signals, image enhancement, medical measurement, etc. If the observed signals are stationary, Gaussian, white ones, then blind separation is essentially impossible. Conversely, blind separation can be realized by exploiting some information on nonstationary, non-Gaussian, or nonwhite characteristics of the observed signals, if any. Most of the conventional methods stipulate that the source signals are non-Gaussian, and use some high-order moments or cumulants. However, it is sometimes difficult to accurately estimate non-Gaussian statistics because random signals in practice are usually not so far from Gaussian. In this paper we propose an approach that utilizes only second-order moments of the observed signals. We consider two cases: (i) the source signals are nonstationary; (ii) the source signals have some temporal correlations, i.e., they are nonwhite signals. To realize signal separation we consider a recovering filter which takes in the observed signals as input and provides an estimate of the source signals as output. The parameters of the filter are determined such that all the outputs of the filter be mutually statistically independent. As a criterion of the statistical independence we adopt the well-known mutual information between the outputs. The adaptation rule for the filter's parameters is derived from the steepest descent minimization of the criterion function. A remarkable feature of our approach is that it is able to treat time-convolutive mixtures of a general number of (stationary) source signals. In contrast, most of the conventional studies on blind separation only consider the static mixing of the source signals. Namely, any delay in the mixing process is not taken into account. So, those methods are useless for many of the important applications of blind separation, e.g., separation of sound signals. Although there are some studies that deal with convolutive mixtures involving some delay, all of them consider only the case of two sources (2 x 2 channels) and do not seem extendible to the case of more than two sources. In our approach, also for convolutive mixtures, the adaptation rule is easily obtained by defining the information criterion in the frequency domain. 6. Dr. Eric MOREAU, and Prof. Odile MACCHI Laboratoire des Signaux et Systems, CNRS-ESE, FRANCE "Adaptive Unsupervised Separation of Discrete Sources" ABSTRACT: We consider the unsupervised source separation problem where observations are captured at the output of an unknown linear mixture of random signals called sources. The sources are assumed discrete, zero-mean and statistically independent. In this paper we consider the problem with a prewhitening stage. The a priori knowledge that sources are discrete with known level, is used in order to improve performances. A novel contrast which combines two parts, is proved. The first part forces statistical independence of the outputs while the second one forces the outputs to have the known distribution. Then a stochastic gradient adaptive algorithm is proposed. Its performance is illustrated thanks to computer simulations that clearly show that the novel contrast achieves much better performance. 7. Dr. Adel BELOUCHRANI, and Prof. Jean-Francois CARDOSO , Telecom Paris, CNRS URA 820, GdR TdSI 46 rue Barrault, 75634 Paris Cedex 13, FRANCE "Maximum Likelihood Source Separation by the Expectation-Maximization Technique: Deterministic and Stochastic Implementation" Abstract: This paper deals with the source separation problem which consists in the separation of a mixture of independent sources without a priori knowledge about the mixing matrix. When the source distributions are known in advance, this problem can be solved via the maximum likelihood (ML) approach by maximizing the data likelihood function using (i) the Expectation-Maximization (EM) algorithm and (ii) a stochastic version of it, the SEM, wich is efficiently implemented by resorting to Metropolis sampler. Two important features of our algorithm are that (a) the covariance of the additive noise can be estimated as a regular parameter, (b) in the case of discrete sources, it is possible to separate more sources than sensors. The effectiveness of this method is illustrated by numerical simulations. 8. Prof. L. TONG and Dr. X. CHEN, University of Connecticut, USA. "Blind Separation of Dynamically Mixed Multiple Sources and its Applications in CDMA Systems" Abstract: In this paper, we consider the problem of separating dynamically mixed multiple sources. Specifically, we address the problem of recovering the sources of an multiple-input multiple-output system. Two issues will be addressed: (i) Source Blind Separability; (ii) Blind Signal Separation Algorithms. Applications of the proposed approach to code-division multiple-access schemes in wireless communication are presented. 9. Prof. Shun-ichi AMARI, Prof. Andrzej CICHOCKI and Dr. Howard Hua YANG, Frontier Research Program RIKEN (Institute of Physical and Chemical Research), Wako-shi, JAPAN "Multi-layer Neural Networks with Local Learning Rules for Blind Separation of Sources" Abstract: In this paper we will propose multi-layer neural network models (feedforward and recurrent) with novel, local, adaptive, unsupervised learning rules which enable not only to separate on-line independent sources but also determine the number of active sources. In other words, we assume that the number of sources and their waveforms are completely unknown. Moreover, the separation problem can be very ill-conditioned and/or badly scaled. In fact the performance of the learning algorithm is independent of scaling factors and a condition number of the mixing matrix. Universal (flexible) computer simulation program will be presented which enable comparison of validity and performance of various recently developed adaptive on-line learning algorithms. 10.. Dr.L. De Lathauwer and Dr.P. Comon; E.E. Dept. - ESAT - SISTA, K.U.Leuven, BELGIUM, CNRS - I3S, Sophia Antipolis, Valbonne, FRANCE "Higher - Order Power Method" Abstract The scientific boom in the field of higher-order statistics involves an increasing need for numerical tools in multi-linear algebra: higher-order moments and cumulants of multivariate stochastic processes are higher-order tensors. We consider the problem of generalizing the computation of the best rank-R approximation of a given matrix to the computation of the best rank-(R1,R2,...,RN) approximation of an Nth-order tensor. We mainly focus on the best rank-1 approximation of third-order tensors. It is shown that this problem leads in a very natural way to a higher-order equivalent of the well-known power method for the computation of the eigendecomposition of matrices. It can be proved that each power iteration step decreases the least-squares error between the initial tensor and the lower-rank estimate. In the tensor case several stationary points might exist, each with a different domain of attraction. Surprisingly, the power iteration for a super-symmetric tensor can produce intermediate results that are unsymmetric. Imposing symmetry on the algorithm does not necessarily improve the convergence speed; the symmetric power iteration can even fail to converge. In the matrix case truncation of the singular value decomposition (SVD) yields the best rank-R approximation; in the tensor case it can only be proved that truncation of the higher-order singular value decomposition (HOSVD) yields a fairly good approximation. All our simulations show that the HOSVD-guess belongs to the attraction region corresponding to the optimal fit. --------------------------------------------------------------------------- From pazzani@super-pan.ICS.UCI.EDU Thu Sep 21 19:06:30 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Thu, 21 Sep 95 19:06:25 -0500; AA22897 Received: from paris.ics.uci.edu by lucy.cs.wisc.edu; Thu, 21 Sep 95 19:06:01 -0500 Received: from super-pan.ics.uci.edu by paris.ics.uci.edu id aa23820; 21 Sep 95 11:52 PDT To: ML-LIST:; Subject: Machine Learning List: Vol. 7, No. 16 Reply-To: ml@ics.uci.edu Date: Thu, 21 Sep 1995 11:15:29 -0700 From: Michael Pazzani Message-Id: <9509211152.aa23820@paris.ics.uci.edu> Machine Learning List: Vol. 7, No. 16 Thursday, September 21, 1995 Contents: Re: Genetic Programming vs A Strawman Algorithm Deadline Extension: special issue of MLJ on ILP Third International Colloquium on Grammatical Inference NIPS*95 Registration Info Available NIPS*95 Workshop on Transfer: Call for Participation FINAL PROGRAM: EPIA'95 Learning Theory Bib update (README) Second and Final CFP: special issue of Evolutionary Computation Workshop on Data Engineering for Inductive Learning Foundations of GAs 1996 - CFP SPECIAL ISSUE of Connection Science Special AI session of SECTAM XVIII ICONIP96 The Machine Learning List is moderated. Contributions should be relevant to the scientific study of machine learning. Mail contributions to ml@ics.uci.edu. Mail requests to be added or deleted to ml-request@ics.uci.edu. Back issues may be FTP'd from ics.uci.edu in pub/ml-list/V/ or N.Z where X and N are the volume and number of the issue; ID: anonymous PASSWORD: URL- http://www.ics.uci.edu/AI/ML/Machine-Learning.html ---------------------------------------------------------------------- From: Barak Pearlmutter Date: Fri, 8 Sep 1995 23:30:57 -0400 Subject: Re: Genetic Programming vs A Strawman Algorithm Koza's comments (ML list 7.14) on Lang's paper (ML95) seemed to rather miss the point. Lang applied a deliberately weak and stupid algorithm (RMHC), which no one in their right mind would ever use, to a tiny problem Koza himself had benchmarked GP on. RMHC was faster. Koza spent the bulk of his comments railing againt RMHC, showing how it is a really really bad algorithm with terrible theoretical properties. But, using Koza's own numbers, GP is demonstrably slower than RMHC on even this teeny tiny problem; and again using Koza's own numbers, GP is demonstrably scaling worse than RMHC in this miniscule toy domain. So arguments about awful RMHC is would seem to apply with a vengeance to GP. Barak Pearlmutter Siemens Corporate Research 755 College Road East Princeton, NJ 08540 bap@scr.siemens.com ------------------------------ From: David.Page@comlab.ox.ac.uk Date: Sat, 2 Sep 95 20:54:17 BST Subject: Deadline Extension: special issue of MLJ on ILP ********************************************************** IMPORTANT UPDATE---DEADLINE EXTENDED TO OCTOBER 15 FOR: ********************************************************** Special Issue: INDUCTIVE LOGIC PROGRAMMING MACHINE LEARNING JOURNAL The deadline for the special issue of Machine Learning Journal on Inductive Logic Programming has been extended to October 15, 1995 (from September 15, 1995). Below is the Call for Papers for this special issue. ********************************************************** CALL FOR PAPERS ********************************************************** Special Issue: INDUCTIVE LOGIC PROGRAMMING MACHINE LEARNING JOURNAL Edited by Stephen Muggleton and David Page Oxford University Computing Laboratory Inductive Logic Programming (ILP) is a growing research area spawned by Machine Learning and Logic Programming. While the in- fluence of Logic Programming has encouraged the development of strong theoretical foundations, the new area has inherited its experimental orientation from Machine Learning. Already ILP has been applied successfully to a variety of complex problems, in- cluding structure-activity and mutagenicity prediction of pharma- ceutical chemicals, protein secondary-structure prediction, finite-element mesh design, and optimal play in chess endgames. For this special issue we encourage submission of papers describ- ing novel ILP algorithms, experimental applications, or theoreti- cal results. Submission deadline: October 15, 1995 It is the editors' intention to publish the special issue as a book as well. Papers should be double spaced and 8,000 to 12,000 words in length, with full-page figures counting for 400 words. All sub- missions will be subject to the standard review procedure. Send three (3) copies of submissions to: David Page Phone: +44-865-283-520 Oxford University Computing Lab Fax: +44-865-273-839 Wolfson Building David.Page@prg.ox.ac.uk Parks Road Oxford, OX1 3QD U. K. Also mail five (5) copies of submitted papers to: Karen Cullen Phone: (617) 871-6300 MACHINE LEARNING Editorial Office karen@world.std.com Kluwer Academic Publishers 101 Philip Drive Norwell, MA 02061 U. S. A. Note: Machine Learning is now accepting submission of final copy in electronic form. A latex style file and related files are available via anonymous ftp from ftp.std.com. Look in Kluwer/styles/journals for the files README, smjrnl.doc, smjrnl.sty, smjsamp.tex, smjtmpl.tex, or smjstyles.tar (which contains them all). ------------------------------ From: MICLET Date: Fri, 8 Sep 95 11:21:57 +0100 Subject: Third International Colloquium on Grammatical Inference Third International Colloquium on Grammatical Inference (ICGI-96) Montpellier (France), September 25-27, 1996 With the help of the Special Interest Group in Natural Language Learning (SIGNLL) of the ACL. Chairperson: Laurent Miclet (IRISA-ENSSAT, Lannion, France) miclet@enssat.fr Organization : Colin de la Higuera (LIRMM, Montpellier, France) cdlh@lirmm.fr Grammatical Inference (GI) is broadly understood as Machine Learning of Grammars and Languages from data. Traditionally, GI has been studied within several contexts: Information Theory, Formal Languages Theory, Computational Linguistics, Machine Learning, Pattern Recognition, Computational Learning Neural Networks, etc. This multidisciplinary perspective, however, has lead so far to a lack of a focused research community. A first attempt to correct this started with the "First Colloquium on Grammatical Inference : Theory, Applications and Alternatives" held in the University of Essex (U.K.), in April 1993. Then followed the "International Colloquium on Grammatical Inference 1994", held in Alicante (Spain), which proceedings have been published by Springer-Verlag as Volume 862 of the Lectures Notes in Artificial Intelligence. =46ollowing these successful meetings, ICGI 96 keeps aiming to provide a forum for discussion of principles, theory and applications of all those aspects of Machine Learning that explicitly focus on Grammars and Languages. Within this framework, topics of interest include, but are not limited to, the following : * Learning Paradigms for Grammars and languages : Cognitive models, Algebraic aspects, Identification in the limit and PAC-Learning ; Stochastic and Corpus-based approaches, Neural Networks, Genetic Algorithms, Fuzzy systems, etc. * Algorithms. * Heuristics. * Benchmarks. * Applications : Natural Language Processing, Language Translation ; Biological Sequences and Time Series Modelization and Prediction ; Image and Speech Recognition, Discrete Events Systems, etc. SCHEDULE : April 1, 1996 Deadline for submitted papers. June 15, 1996 Notification of acceptance and referrees comments. July 15, 1996 Camera ready copy. September 25-27, 1996 Colloquium. Please submit (not via electronic mail) three copies of your full length art= icle (maximum 12 pages, 12 pt. font, including figures, tables, references, etc.) to : L. Miclet IRISA-ENSSAT BP 447 - 6, Rue de K=E9rampont 22305 LANNION Cedex FRANCE The Proceedings of the Colloquium will be considered for publication as a volume in the Springer-Verlag Lecture Notes Series in Artificial Intelligence. Information on ICGI'96 is on the www page : http://itkwww.kub.nl:2080/itk/Docs/Projects/Walter/icgi.html ------------------------------ From: David Cohn Date: Mon, 11 Sep 95 18:06:01 EDT Subject: NIPS*95 Registration Info Available CONFERENCE ANNOUNCEMENT Neural Information Processing Systems Natural and Synthetic Monday, Nov. 27 - Saturday, Dec. 2, 1995 Denver, Colorado http://www.cs.cmu.edu/Web/Groups/NIPS/NIPS.html This is the ninth meeting of an interdisciplinary conference which brings together neuroscientists, engineers, computer scientists, cognitive scientists, physicists, and mathematicians interested in all aspects of neural processing and computation. The confer- ence will include invited talks, and oral and poster presenta- tions of refereed papers. There will be no parallel sessions. There will also be one day of tutorial presentations (Nov. 27) preceding the regular session, and two days of focused workshops will follow at a nearby ski area (Dec. 1-2). Major conference topics include: Neuroscience, Theory, Implemen- tations, Applications, Algorithms & Architectures, Visual Pro- cessing, Speech/Handwriting/Signal Processing, Cognitive Science & AI, Control, Navigation and Planning. Detailed information and registration materials are available electronically at http://www.cs.cmu.edu/Web/Groups/NIPS/NIPS.html ftp://psyche.mit.edu/pub/NIPS95/ Students who require financial support to attend the conference are urged to retrieve a copy of the registration brochure as soon as possible in order to meet the aid application deadline. Mail general inquiries/requests for registration material to: NIPS*95 Registration Dept. of Mathematical and Computer Sciences Colorado School of Mines Golden, CO 80401 USA FAX: (303) 273-3875 e-mail: nips95@mines.colorado.edu ------------------------------ From: Rich Caruana Date: Fri, 15 Sep 95 09:26:38 -0400 Subject: NIPS*95 Workshop on Transfer: Call for Participation *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* *-* POST-NIPS*95 WORKSHOP *-* *-* December 1-2, 1995 *-* *-* Vail, Colorado *-* *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* *-* CALL FOR PARTICIPATION *-* *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-* TITLE: "Learning to Learn: Knowledge Consolidation and Transfer in Inductive Systems" ORGANIZERS: Jon Baxter, Rich Caruana, Tom Mitchell, Lori Pratt, Danny Silver, Sebastian Thrun. INVITED TALKS BY: Leo Breiman (Stanford, undecided) Tom Mitchell (CMU) Tomaso Poggio (MIT) Noel Sharkey (Sheffield) Jude Shavlik (Wisconsin) WEB PAGES (for more information): Our Workshop: http://www.cs.cmu.edu/afs/cs/usr/caruana/pub/transfer.html NIPS*95 Info: http://www.cs.cmu.edu/afs/cs/project/cnbc/nips/NIPS.html WORKSHOP DESCRIPTION: The power of tabula rasa learning is limited. Because of this, interest is increasing in methods that capitalize on previously acquired domain knowledge. Examples of these methods include: o using symbolic domain theories to bias connectionist networks o using unsupervised learning on a large corpus of unlabelled data to learn features useful for subsequent supervised learning on a smaller labelled corpus o using models previously learned for other problems as a bias when learning new, but related, problems o using extra outputs on a connectionist network to bias the hidden layer representation towards more predictive features There are many different approaches: hints, knowledge-based artificial neural nets (KBANN), explanation-based neural nets (EBNN), multitask learning (MTL), knowledge consolidation, etc. What they all have in common is the attempt to transfer knowledge from other sources to benefit the current inductive task. The goal of this workshop is to provide an opportunity for researchers and practitioners to discuss problems and progress in knowledge transfer in learning. We hope to identify research directions, debate different theories and approaches, discover unifying principles, and begin to start answering questions like: o when will transfer help -- or hinder? o what should be transferred? o how should it be transferred? o what are the benefits? o in what domains is transfer most useful? SUBMISSIONS: We solicit presentations from anyone working in (or near): o Sequential/incremental, compositional (learning by parts), and parallel learning o Task knowledge transfer (symbolic-neural, neural-neural) o Adaptation of learning algorithms based on prior learning o Learning domain-specific inductive bias o Combining predictions made for related tasks from one domain o Combining supervised learning (where the goal is to learn one feature from the other features) with unsupervised learning (where the goal is to learn every feature from all the other features) o Combining symbolic and connectionist methods via transfer o Fundamental problems/issues in learning to learn o Theoretical models of learning to learn o Cognitive models of, or evidence for, transfer in learning Please send a short (one page or less) description of what you want to present to one of the co-chairs below by Oct 15. Email is preferred. We'll select from the submissions and publish a workshop schedule by Nov 1. Preference will be given to submissions that are likely to generate debate and that go beyond summarizing prior published work by raising important issues or suggesting directions for future work. Suggestions for moderator or panel-led discussions (e.g., sequential vs. parallel transfer) are also encouraged. We plan to run the workshop as a workshop, not as a mini conference, so be daring! We look forward to your submission. Rich Caruana Daniel L. Silver School of Computer Science Department of Computer Science Carnegie Mellon University Middlesex College 5000 Forbes Avenue University of Western Ontario Pittsburgh, PA 15213, USA London, Ontario, Canada N6A 3K7 email: caruana@cs.cmu.edu email: dsilver@csd.uwo.ca ph: (412) 268-3043 ph: (519) 473-6168 fax: (412) 268-5576 fax: (519) 661-3515 ------------------------------ From: Nuno Joao Mamede Date: Sun, 17 Sep 95 20:26:45 +0100 Subject: FINAL PROGRAM: EPIA'95 X-Mts: smtp EPIA'95 - PRELIMINARY PROGRAM SEVENTH PORTUGUESE CONFERENCE ON ARTIFICIAL INTELLIGENCE Casino Park Hotel, Funchal, Madeira Island, Portugal 3-6 October, 1995 (Under the auspices of the Portuguese Association for AI) The 7th Portuguese Conference on Artificial Intelligence will be held at Funchal, Madeira Island, Portugal, October 3-6, 1995. As in the past, EPIA 95 is an international conference with English as the official language. The conference covers all areas of Artificial Intelligence, including theoretical areas, foundational areas, and applications. The scientific program consists of invited lectures, tutorials, demonstrations and paper presentations. There will also be parallel workshops on Expert Systems, Fuzzy Logic and Neural Networks, and Applications of AI to Robotics and Vision Systems. ====================================================================== FINAL PROGRAM (Summary) ====================================================================== Tuesday - October, 3 95 ~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 12:30 TUTORIAL 1 - Artificial Life and Autonomous Robots Luc Steels 9:00 - 12:30 TUTORIAL 3 - Introduction to Artificial Intelligence Ernesto Costa (in Portuguese) 14:30 - 18:00 TUTORIAL 2 - Virtual Reality - The AI perspective David Hogg 14:30 - 18:00 TUTORIAL 4 - Design of Expert Systems Ernesto Morgado (in Portuguese) Wednsday - October, 4 95 ~~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 9:30 OPENING SESSION 9:40 - 10:30 QUALITATIVE REASONING 10:30 - 10:50 Coffee break 10:50 - 12:30 NEURAL NETWORKS & DISTRIBUTED ARTIFICIAL INTELLIGENCE 10:50 - 12:30 FUZZY LOGIC & NEURAL NETWORKS WORKSHOP 12:30 - 14:00 Lunch 14:00 - 15:30 Invited Lecture by LUIS B. ALMEIDA (IST - Portugal) "The Connectionist Paradigm and AI" 15:30 - 15:50 Coffee break 15:50 - 18:00 BELIEF REVISION & NON-MONOTONIC REASONING 15:50 - 18:30 FUZZY LOGIC & NEURAL NETWORKS WORKSHOP 15:50 - 18:30 APPLICATIONS OF EXPERT SYSTEMS WORKSHOP 20:00 - Welcome Dinner (With a performance of the Univ. of Madeira "tuna") Thursday - October, 5 95 ~~~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 10:30 Invited Lecture by RODNEY BROOKS (MIT - USA) "The Evolutionist Approach - Past, Present and Future of AI" 10:30 - 10:50 Coffee break 10:50 - 12:30 APPLICATIONS OF EXPERT SYSTEMS WORKSHOP 10:50 - 11:40 ROBOTICS AND CONTROL 11:40 - 12:30 POSTER SECTION 12:30 - 14:00 Lunch 14:00 - 15:15 MACHINE LEARNING 15:15 - 15:35 Coffee break 15:35 - 17:05 Invited Lecture by MARVIN MINSKY (MIT - USA) "Why Human Brains Can't Really Think" 17:15 - 18:30 Visit to the Madeira Wine Cellars Friday - October, 6 95 ~~~~~~~~~~~~~~~~~~~~~~ 9:00 - 10:30 INVITED LECTURE by Manuela Veloso (CMU - USA) "Planning and Learning in Intelligent Agents" 10:30 - 10:50 Coffee break 10:50 - 12:30 STREAM 1: PLANNING AND CASE-BASED REASONING 10:50 - 12:30 STREAM 2: CONSTRAINT-BASED REASONING 10:50 - 12:30 APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP 12:30 - 14:00 Lunch 14:00 - 15:30 STREAM 1: AUTOMATED REASONING AND THEOREM PROVING 14:00 - 15:30 STREAM 2: GENETIC ALGORITHMS & THEORY OF COMPUTATION 14:00 - 15:30 APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP 15:30 - 15:50 Coffee break 15:50 - 17:30 PANNEL (The Next Frontiers of AI: the Role of Foundations) 18:00 - 19:00 APPIA meeting 20:00 Farewell Dinner (with folklore dances show) Saturday - October, 7 95 ~~~~~~~~~~~~~~~~~~~~~~~~ TOUR 1 - Island Tour (full day) TOUR 2 - Ribeiro Frio/Portela Walking Tour (full day) TOUR 3 - Eira do Serrado (half day) ====================================================================== PAPERS TO BE PRESENTED IN EACH SESSION ====================================================================== AUTOMATED REASONING AND THEOREM PROVING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Terminological Meta-Reasoning by Reification and Multiple Contexts Klemens Schnattinger, Udo Hahn, Manfred Klenner CLIF, Freiburg University, Germany A New Continuous Propositional Logic Riccardo Poli, Mark Ryan, Aaron Sloman SCS, The University of Birmingham, UK Super-Polynomial Speed-Ups in Proof Length by New Tautologies Uwe Egly FG Intellektik, TH Darmstadt, Germany BELIEF REVISION ~~~~~~~~~~~~~~~ Belief Revision in Non-Monotonic Reasoning Jose Alferes, Luis Moniz Pereira, T. Przymusinski DM, U. Evora, and CRIA, U. Nova de Lisboa, Portugal, and University of California at Riverside, USA A New Representation of JTMS Truong Quoc Dung IRIDIA, Universite Libre de Bruxelles, Belgium CONSTRAINT-BASED REASONING ~~~~~~~~~~~~~~~~~~~~~~~~~~ The Retrieval Problem in a Concept Language with Number Restrictions Aida Vitoria, Margarida Mamede, Luis Monteiro DI, Universidade Nova de Lisboa, Portugal Formalizing Local Propagation in Constraint Maintenance Systems Gilles Trombettoni INRIA-CERMICS, France A Dependency Parser of Korean Based on Connectionist/Symbolic Techniques Jong-Hyeok Lee, Geunbae Lee Pohang University of Science and Technology, Korea A Symbiotic Approach to Arc and Path Consistency Checking Pierre Berlandier INRIA-CERMICS, France DISTRIBUTED ARTIFICIAL INTELLIGENCE ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Where Do Intentions Come From?: A Framework for Goals and Intentions Adoption, Derivation and Evolution Graca Gaspar, Helder Coelho Faculdade de Ciencias de Lisboa, and INESC, Portugal A Closer Look to Artificial Learning Environments Helder Coelho, Augusto Eusebio, Ernesto Costa INESC, Portugal, and DEI, Universidade de Coimbra, Portugal Building Multi-Agent Societies from Descriptions to Systems: Inter-Layer Translations Helder Coelho, Luis Antunes, Luis Moniz INESC, Portugal GENETIC ALGORITHMS ~~~~~~~~~~~~~~~~~~ GA/TS: A Hybrid Approach for Job Shop Scheduling in a Production System Jose Ramon Zubizarreta, Javier Arrieta Facultad de Informatica de San Sebastian, Spain MACHINE LEARNING ~~~~~~~~~~~~~~~~ A Controlled Experiment: Evolution for Learning Difficult Image Classification Astro Teller, Manuela Veloso Carnegie Mellon University, USA Minimal Model Complexity Search Chris McConnell CMU School of Computer Science, USA Characterization of Classification Algorithms Joao Gama, Pavel Brazdil LIACC, Universidade do Porto, Portugal NEURAL NETWORKS ~~~~~~~~~~~~~~~ Neurons, Glia and the Borderline Between Subsymbolic and Symbolic Processing J. G. Wallace, K. Bluff Swinburne University of Technology, Australia NON-MONOTONIC REASONING ~~~~~~~~~~~~~~~~~~~~~~~ Arguments and Defeat in Argument-Based Nonmonotonic Reasoning Bart Verheij University of Limburg, The Netherlands A Preference Semantics for Ground Nonmonotonic Modal Logics Daniele Nardi, Riccardo Rosati DIS, Universita di Roma ``la Sapienza", Italy Logical Omniscience vs. Logical Ignorance On a Dilemma of Epistemic Logic Ho Ngoc Duc ILPS, University of Leipzig, Germany PLANNING AND CASE-BASED REASONING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ On the Role of Splitting and Merging Past Cases for Generation of New Solutions Carlos Bento, Penousal Machado, Ernesto Costa DEI, Universidade de Coimbra, Portugal Theorem Proving by Analogy - A Compelling Example Erica Melis Department of AI, University of Edimburgh, Scotland Non-Atomic Actions in the Situation Calculus Jose Julio Alferes, Renwei Li, Luis Moniz Pereira CRIA and DCS, Universidade Nova de Lisboa, Portugal Planning under Uncertainty: A Qualitative Approach Nikos Karacapilidis FIT.KI, GMD, Sankt Augustin, Germany QUALITATIVE REASONING ~~~~~~~~~~~~~~~~~~~~~ Qualitative Reasoning under Uncertainty Daniel Pacholczyk DMI, U.F.R., Science d'Angers, France Systematic Construction of Qualitative Physics-Based Rules for Process Diagnostics Jaques Reifman, Thomas Y.C. Wei Argonne National Laboratory, USA ROBOTICS AND CONTROL ~~~~~~~~~~~~~~~~~~~~ Integrated Process Supervision (IPS): A Structured Approach to Expert Control Chai Quek, P.W. Ng, M. Pasquier Nanyang Technological University, Singapore Using Stochastic Grammars to Learn Robotic Tasks Pedro Lima, George Saridis ISR, Technical Univ. of Lisbon, Portugal, and Rensselaer Polytechnic Institute, USA THEORY OF COMPUTATION ~~~~~~~~~~~~~~~~~~~~~ Constraint Categorial Grammars Luis Damas, Nelma Moreira LIACC, Universidade do Porto, Portugal A New Translation Algorithm from Lambda Calculus into Combinatory Logic Sabine Broda, Luis Damas LIACC, Universidade do Porto, Portugal POSTER SECTION ~~~~~~~~~~~~~~ Interlocking Multi-Agent and Blackboard Architectures Bernhard Kipper DCS, University of Saarbrucken, Germany A Model Theory for Paraconsistent Logic Programming Carlos Viegas Damasio, Luis Moniz Pereira CRIA, and DCS, Universidade Nova de Lisboa, Portugal Promoting Software Reuse Through Explicit Knowledge Representation Carmen Fernandez-Chamizo, Pedro A. Gonzalez-Calero, Mercedes Gomez-Albarran Universidad Complutense, Spain Efficient Learning in Multi-Layered Perceptron Using the Grow-And-Learn Algorithm Gildas Cherruel, Bassel Solaiman, Yvon Autret Univ. de Bretagne Occidentale, and TNI, and ENSTB, France An Non-Diffident Combinatorial Optimization Algorithm Gilles Trombettoni, Bertrand Neveu INRIA-CERMICS, France Modelling Diagnosis Systems with Logic Programming Iara Mora, Jose Alferes CRIA, U. Nova de Lisboa, and DM, U. Evora, Portugal Agreement: A Logical Approach to Approximate Reasoning Luis Custodio, Carlos Pinto-Ferreira ISR, Technical University of Lisbon, Portugal Constructing Extensions by Resolving a System of Linear Equations Messaoudi Nadia Universite Aix-Marseille II, France Presenting Significant Information in Expert System Explanation Michael Wolverton Daresbury Rutherford Appleton Laboratory, UK A Cognitive Model of Problem Solving with Incomplete Information Nathalie Chaignaud LIPN, Universite Paris-Nord, France Filtering Software Specifications Written In Natural Language Nuria Castell, Angels Hernandez Universitat Politecnica de Catalunya, Spain Parsimonious Diagnosis in SNePS Pedro A. Matos, Joao P. Martins DEM, Technical University of Lisbon, Portugal Syntactic and Semantic Filtering in a Chart Parser Sayan Bhattacharyya, Steven L. Lytinen University of Michigan, and DePaul University, USA GA Approach to solving Multiple Vehicle Routing Problem Slavko Krajcar, Davor Skrlec, Branko Pribicevic, Snjezana Blagajac Faculty of Electrical Eng. and Computing, Croatia Multilevel Refinement Planning in an Interval-Based Temporal Logic Werner Stephan and Susanne Biundo German Research Center for AI, Germany APPLICATIONS OF EXPERT SYSTEMS WORKSHOP ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Wednesday, 4 15:50 - 16:15 CGD - An Expert System for Loan Analysis Decision Vasco Moreira, Andre Frazao, Elisabete Silva, Ernesto Costa Caixa Geral de Depositos, Dir. Organizacao Informatica Lisboa, Portugal 16:15 - 16:40 A Quantitative Method for Performing A Cost-Benefit Analysis of Expert System Projects Ramu Kannan, Reza Khorramshahgol, Mohan Tanniru Dept. of Management Science and Economics, Coppin State College, Baltimore, USA 16:40 - 17:05 DARE: a Knowledge-Based System for the Diagnosis of Neuromuscular Disorders J. Cruz, P. Barahona, A. P. Figueiredo, M. Veloso, M. Carvalho UNINOVA, Portugal 17:05 - 17:30 A Cooperative Multi-Agent System for Strategic Decision Making Suzanne Pinson Jorge Louca Universite Paris IX - Dauphine, Paris, France 17:30 - 17:55 A Hybrid Model for Classification Expert Systems Sergio Rosa, Beatriz Leao Instituto de Informatica UFRGS, Porto Alegre, Brasil Thursday, 5 10:50 - 11:15 PERMEX - Expert System for Corrosion Failure Analysis Fernando Lopes, A. Novais, N. Mamede, C. Rangel INETI, DMS, Lisboa, Portugal 11:15 - 11:40 Architectural Aspects of an Intelligent DSS for Flow Shop Production Control Ioannis Hatzilygeroudis, D. Sofotassios, N. Dendris, P. Spirakis, A. Tsakalidis Dept. of Computer Engin. and Informatics, Univ. of Patras, Greece 11:40 - 12:05 Advances in Explanation Facilities for Expert Systems Keith Darlington School of Computing, Information Systems and Mathematics South Bank University, London, UK 12:05 - 12:30 A Deferred Communication in a Parallel Distributed Expert System Shell Wided Lejouad SECOIA Project, Sophia Antipolis, France APPLICATIONS OF AI TO ROBOTICS AND VISION SYSTEMS WORKSHOP ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Friday, 6 10:50 - 11:15 Designing and Implementing Real Walking Agents Using Virtual Environments Aleix Martinez Universitat Autonoma de Barcelona, Dept. Informatica, Spain 11:15 - 11:40 Multi-Layer Perceptrons for Task Visual Servoing in Robotics Nadine Rondel, Gilles Burel Thomson CSF-LER, France 11:40 - 12:05 Heuristic Autonomous Mobile Robot Using Visual Servoing Jean-Charles Bonin, Fernandoo De Carvalho Gomes Laboratorio de Inteligencia Artificial-LIA, Fortaleza, Brasil 12:05 - 12:30 An Integrated Approach to Position a Robot Arm in a System for Planar Part Grasping Pedro Sanz, Juan Domingo Universitat Jaume I, Dpto. Informatica, Castellon, Spain 14:00 - 14:25 Learning and Recall of Robot Manipulator Motions Using Driver Programs Frank Smieja, Uwe Bayer GMD, Schloss Birlinghoven, Germany 14:25 - 14:50 Selective Visual Perception Driven by Cues from Speech Processing Reinhard Moratz AG Angewandte Informatikj, Universitaet Bielefeld, Germany 14:50 - 15:15 Autonomous Robots and Active Vision Systems: Issues on Architectures an Integration Helder Araujo, Jorge Dias, Jorge Batista, Paulo Peixoto ISR-Coimbra, Universidade de Coimbra, Portugal 15:15 - 15:35 Learning from Perception, Success and Failure in a Team of Autonomous Mobile Robots Arvin Agah, George Bekey Inst. for Robotics and Intell.Syst., Univ. of Southern California, USA FUZZY LOGIC & NEURAL NETWORKS WORKSHOP IN ENGINEERING ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Wednesday, 4 - Session 1 10:50 - 11:15 A Fuzzy Logic Controller for Supraconductivity Measuring N. Zimic, J. Ficzko, M. Mraz, J. Virant Faculty of Electrical & Computer Engineering Science University of Ljubljana, Slovenia 11:15 - 11:40 Complex Data and Fuzziness in Database Applications Adnan Yazici Dept. of Computer Engineering, Middle East Technical University, Turkey 11:40 - 12:05 Car License Plate Recognition with Neural Networks and Fuzzy Logic J. Nijhuis, et.al. Dept. of Computer Science, Groningen University, The Netherlands 12:05 - 12:30 Similarity-Based Self-organized Clustering Jurgen Rahmel Center for Learning Systems & Applications, University of Kaiserslautern, Germany 15:50 - 16:15 On the Representation of Data for Optimal Learning M. Brugge, J. Nijhuis, W. Jansen, H. Drenth, L. Spaanenburg Dept. of Computer Science, Groningen University, The Netherlands 16:15 - 16:40 Artificial Neural Net-Based Controllers for Real Process Control Petr Pivonka, Jan Zizka Dept. of Automatic Control and Instrumentation Technical University of Brno, Czech Republic 16:40 - 17:05 A Production Line for Generating Clinical Decision Support Systems Patrik Eklund Dept. of Computing Science, Umea University, Sweden 17:05 - 17:30 Knowledge Discovery Using Hierarchical Connectionist Marie Pai, Robin Ying AT&T Bell Laboratories, USA 17:30 - 17:55 Growing Filters for Finite Impulse Response Networks M. Diepenhorst, J. Nijhuis, R. Venema, L. Spaanenburg Dept. of Computer Science, Groningen University, The Netherlands ====================================================================== ENQUIRIES ADDRESS ====================================================================== EPIA'95 - INESC E-mail: epia95@inesc.pt Av. Alves Redol, 9 Fax: 351-1-525843 1000 Lisboa Voice: 351-1-3100325 PORTUGAL Home Page: http://www.isr.ist.utl.pt/~cpf/epia95 ====================================================================== SUPPORTERS ====================================================================== Banco Nacional Ultramarino Governo Regional da Madeira Instituto Superior Tecnico SISCOG - Sistemas Cognitivos INESC CITMA IBM TAPair Portugal ------------------------------ From: William Ian Gasarch Date: Mon, 18 Sep 1995 12:30:52 -0400 Subject: Learning Theory Bib update (README) This is the README file for using the Computational Learning Theory bibliography. This message includes the original statement about history /purpose/ how to use, as well as (short) items that have been added whenever I updated it. This is being sent in September 1995 1) From now on we will try to get you a fresh version of this out to you between COLT submissions and COLT final versions being due. This way the final versions can have up-to-date references. 2) When you email me entries (see below for how) make sure they go through bibtex at your home site. This will save me time weeding out errors. In particular math things should be in math mode. Also \uz is not a command. 3) Do not put in comments like ``this supercedes ag-ersu-90'' better to have a comment like ``Earlier version appeared in 1st COLT, 1988'' The reason for this is that having `ag-ersu-90' appear in a journal in a bibliography is meaningless, while having `COLT 1988' is meaningful. 4) if a name has an accent, like Erd\"os it should be written Erd{\"o}s in the file. 5) Use `Lecture Notes in Computer Science' or `Lecture Notes in Artificial Intelligence' rather than LNCS and LNAI. If the article is in such a publication please include this information and what volume. 6) There is a website that mirrors this bibliography as well as many others in computer science. Ours is under AI The website is http://liinwww.ira.uka.de/bibliography/index.html This is being sent in July 1995. The next one will be sent shortly after COLT, in August 1995. In the future it should come out once every three months. One of the times it will come out is when everyone is preparing their COLT papers (final versions). This is being sent in Late September 1994 The reason I am sending this update ahead of schedule is because there are some changes: 1) Information about them is in a file called bibinfo which is part of the package you can access OR you can just ftp that file separately. 2) We are using a program bibclean in our software. It made our software more maintainable and shorter but should have no affect on the user. 3) We are no longer supplying coltbib.dvi.Z We are instead supplying coltbib.tex from which you can create coltbib.dvi.Z. dvi files do not email that well. If enough people object to this change let me know and I'll change it back. 4) I've updated the list of available files in this README file. Its in the section called WHAT FILES ARE AVAILABLE? REQUEST: PLEASE check papers that YOU wrote that may require updated references. Send updates by just submitting new entries (I will take care of duplicates) I would like updates by OCTOBER 30 to put into the next version. PLEASE look at the file LOOKUP and if you have any info to clarify those entries please let me know. HISTORY AND OVERVIEW This bibliography, an ongoing project whose goal is to offer a reasonably complete database of publications in computational learning theory, came into existence in 1993 after the publication of the Computational Geometry Column in the Spring 1993 SIGACT News, describing the bibligraphy maintained by that community. At the suggestion of Lenny Pitt, Bill Gasarch and Anselm Blumer volunteered to begin a similar effort for computational learning theory. Dana Angluin, Sanjay Jain, Phil Long and Ron Rivest provided invaluable help by allowing their existing online bibliographies to be merged to provide a starting point. The tables of contents from all the COLT conferences were then added, as well as those papers from STOC and FOCS which seemed the most relevant. Gisli Hjaltason provided invaluable help by programming awk scripts to put everything into a standard format and find duplicates. WHAT FILES ARE AVAILABLE? Several files are available via anonymous ftp or email. (Next section says HOW to access them) The files available are the following. 1) colt.bib.Z. This is a compressed bibtex file. If you do decompress it (command: uncompress colt.bib.Z) then you will have a file colt.bib which is a latex biblio file. 2) coltbib.tex This is a short tex file. If you have the colt bib file in a file named colt.bib, and you latex coltbib.tex then you will get a file that, if printed out, would have EVERY ENTRY as they would appear in a bibliography. You probably do not want to print this out, but you might want to preview it. 3) coltbib.ps.Z This is a compressed file that when decompressed is a postscript file that, when printed out, gives you all the entries as they would appear in a bibliography. 4) authority This file contains the conventions that we use in naming conferences and journals in the bibliography file. (e.g. FOCS conference is referred to by Proc. Nth Annu. IEEE Sympos. Found. Comput. Sci.) 5) This README file. 6) coltall One file that when executed produces the five files above. To execute the file type sh coltall or type coltall 7) SOFT is a directory that contains our software and misc files. There is no real need to look in this directory and it is not organized for public use. Its only there to make the job of maintaining this bib easy. 8) pick- a tool for looking through the bib 9) tools- a file with some advise on where to find more sophisticated tools. 10) bibinfo- a file with some advise to find even more sophisticated tools that are becoming standard. 11) LOOKUP- a file of entries, in pairs, that I am trying to find more about so I can better classify them. The file you will be working with the most is colt.bib You are encouraged to use it actively, make additions and corrections, and send the changes to coltbib@cs.umd.edu for integration into a new revision. After unpacking the file you should preserve the original biblio Source file of colt.bib in some read-only form (called oldcolt.bib, say) and make a writeable copy for your actual changes. Later on, you should email me your changes by emailing me either entries that are not in the file or updated versions of entries that are in the file. (Details in next section.) HOW TO ACCESS You can access these files two ways 1) Using ftp Establish an anonymous ftp connection to cs.umd.edu, then change directories to pub/coltbib, then retrieve colt.bib.Z with ftp in binary mode. (This site is on Eastern Standard Time and although we place no restrictions on access time, the hours at which you can retrieve files efficiently may vary with the load on intervening networks.) $ ftp cs.umd.edu Name: anonymous Password: yourname@yoursite ftp> cd pub/coltbib ftp> dir total 924 -rw-r--r-- 1 gasarch 28679 Dec 10 17:57 README drwxr-xr-x 2 gasarch 1024 Dec 10 17:56 SOFT -rw-r--r-- 1 gasarch 7342 Nov 19 15:14 authority -rw-r--r-- 1 gasarch 87055 Nov 19 15:14 colt.bib.Z -rwxr-xr-x 1 gasarch 518953 Nov 19 15:17 coltall -rw-r--r-- 1 gasarch 112879 Nov 19 15:14 coltbib.dvi.Z -rw-r--r-- 1 gasarch 149048 Nov 4 15:13 coltbib.ps.Z ftp> binary ftp> get colt.bib.Z ftp> quit (you could get any of these files this way) Unpack it with uncompress colt.bib.Z If you get the file coltall then put it into an empty directory and execute it. It will produce all the files in compressed form. 2) Using email. If you mail to coltbib@cs.umd.edu then you will get the file coltall in return. If you want all five files or if you cannot ftp then this is the best way to get them. If you just want colt.bib then ftp is better to use. WARNING: Standard bibtex comes with a limit of 750 entries per bibliography, which is laughably small for us. People will be able to produce hardcopy according to their own tastes only once they get their local bibtex reconfigured. HOW TO SUBMIT Email new entries or updated versions of old entries to coltadd@cs.umd.edu They will be intergrated into the next version. What forms are acceptable for entries is discussed below. We will be updating this bibliography on a regular schedule (perhaps every four months). We will email to the colt mailing list a short while before the deadline that we are updating soon, though updates can be sent anytime. BIBLIOGRAPHERS This bibliography is and always has been a product of the good will of volunteers from our community. An essential idea of this project is that it ought to be a social effort, and that everyone who is using this bibliography should feel willing to contribute his or her share in improving it. The following are the volunteers who have helped produce this electronic bibliography, whether in getting it up and running, or converted to bibtex, or continuing to improve its coverage and accuracy. Dana Angluin, Anselm Blumer, Bill Gasarch, Gisli Hjaltason, Sanjay Jain, Bill Jones, Phil Long, Joseph O'Rourke, Ron Rivest However, if the project is to continue, it can only do so through widespread participation. We ask that, if you are using this bibliography, find it helpful, and wish it to carry on, you ``pay forward'' a fraction of the time which it saves you by joining us and contributing updates as described herein. CREATING ENTRIES What goes in? Papers relevant to computational learning theory, which for us means the study of the computational complexity of well-defined learning problems. Thus we are talking algorithms, data structures, analysis of time and storage, lower and upper bounds, etc. We interpret relevance in a rather broad sense, although we prefer that references from cognate areas (such as statistics, recursion theory, psychology of learning, etc.) emphasize books or survey articles rather than individual papers, and that these be included only if they would be referenced by several different papers in the bibliography. In the end, your judgement as a working researcher decides what is relevant and worth inclusion. (A pragmatic test: have you cited or would you cite the item in your own papers?) Future maintenance is easiest if you include only papers which are "stable"; i.e. published and openly available at least in the form of a numbered techreport, and preferably in a conference proceedings. However, it is okay to include preprints too. If the paper is slated to appear somewhere else, that information can usefully annotate the entry for an existing appearance. Mary-Claire van Leunen's book _A Handbook for Scholars_ (Knopf, New York, 1979) suggests that for utmost scholarship nothing short of the original title page should be trusted: "To write a reference, you must have the work you're referring to in front of you.... The temptation to write a reference without having the work before you will be powerful. Resist it. A vague recollection is worthless; a vivid recollection is probably the result of your imagination --- ingenious, no doubt, but of little use to your reader. Don't rely on your memory.... If you must not rely on your own memory, even less should you rely on someone else's. If your only access to a reference is through a secondary source, then you must refer to the secondary source as well as the primary one." We are less concerned with the sheer volume of what you add to the bibliography than with its accuracy and relevance. But please bear in mind that there is a minimum overhead of at least an hour to process each submission in the merging process, making larger submissions more efficient than tiny ones. Coordinating your changes with those of other colleagues, grad students, and so on at your site before sending is greatly appreciated. We are always open to suggestions on how to capture data with best efficiency and least overlap, but at the moment would suggest the following approach: 1) use the bibliography as a bibtex database when typesetting references for your own papers, so that adding entries and making corrections can happen as a natural side effect of your own work. 2) during that process, you will likely wind up referring to papers from some conference or journal year which isn't known to have been covered by the bibliography (see the list below). It would be very helpful if you took the time for at least one such paper to check through the whole volume and ensure that all relevant learning papers have been incorporated in the bibliography. (This doesn't take so long as you might think: by keeping an entry template with repetitive details ready in your editor, within an hour you can enter a full conference of about 50 papers.) 3) please look carefully at entries for papers written by you, or by people at your institution, to ensure their correctness. No one else can do this more accurately or more efficiently. 4) if you are caught up with current events and looking for a pastime, you can work on something from our open problems list; or you can check back through unexamined years of a journal or conference to ensure that all relevant papers are included, correct, and keywordized. FORMATTING ENTRIES Because of the distributed nature of updates, it seems desirable to have some written guidelines for the format of entries, in order that the final product have a consistent style. The following suggestions are based on common practice where discernible, established authorities where possible, and personal opinion where unavoidable. You will likely find existing entries in disagreement with these guidelines. Either the entry or the guidelines should be fixed. If some entry can't be decently handled by the current guidelines, or you think they're just plain wrong in any case, please let us know about it. In the hope of keeping future input work reasonably simple and error-free, a few lexical conventions were set at the time of bibtex conversion, as follows. Where possible you should use lower case (for simplicity), a leading comma ``, volume = 12'' (to make missing commas obvious), and put all text for a field on a single line (to avoid spending time on prettyprinting). If you must break lines, such as in the abstract= or annote= or comments= or note= fields, start subsequent lines with a tab. Special characters and diacriticals should be entered as specified on p.52 of the TeXbook. The common single-letter ones are described below. \' acute sup{\'e}rieur {\'}O'D{\'}unlaing [\*' in troff -ms] \` grave probl{\`e}me Bruy{\`e}re [\*` in troff -ms] \^ circumflex m{\^e}me Lema{\^i}tre [\*' in troff -ms] \~ tilde ma{\~n}ana N{\'u}{\~n}ez [\*' in troff -ms] \v hacek h{\'a}{\v c}ek Matou{\v s}ek [\*C in troff -ms] \c cedilla fran{\c c}ais {\'}Swi{\c a}tek [\*, in troff -ms] \" umlaut f{\"u}r G{\"u}ting [\*: in troff -ms] \H Hung. umlaut Erd{\H o}s Note that diacriticals precede the letter affected. A complication is that in TeX, control sequences specified using letters must somehow be separated from the ordinary letters that follow. A simple way is to use spaces as in "Erd\H os", but this will look like two separate words to bibtex. Another is to use braces as in "Erd\H{o}s", but this too is confounded by bibtex, which (1) normally wants to decapitalize text in titles not protected by braces, to support variant capitalization styles, and (2) will interpret an umlaut \" as the end of a quoted string, unless specially protected. Initially it might seem enough to put braces around the whole word when it contains either a fussy diacritical or (in a title field) a capital letter. However, it turns out that, because of how it handles the author field, bibtex dictates the convention to follow. Since adding a feature to recognize and handle accented characters in author fields (for benefit of the alpha bibliography styles), bibtex requires that we "place the entire accented character in braces; in this case either {\"o} or {\"{o}} will do .... furthermore these braces must not themselves be enclosed in braces (other than the ones that might delimit the entire field or the entire entry), and there must be a backslash as the very first character inside the braces". Thus you should use, for example, {\'O}'D{\'u}nlaing, Matou{\v s}ek, G{\"u}ting, and Erd{\H o}s, and we recommend that for consistency you treat all accents this way in whatever bibtex fields they appear. However you will further have to embrace the whole of any capitalized name that appears in a title field. Such is life with bibtex. Mathematical expressions, including numbers in titles, should always be entered in TeX notation. Author, title, and page information from other than the title page of the paper itself is untrustworthy: you might want to do data entry from a proceedings table of contents for speed, but please take time to proofread against title pages for accuracy. Below is a quick naming of parts for entries in the database, with discussions of the conventions that have evolved. More detailed information on entry formats can be found in the bibtex documentation. Entry type: we ignore some of the fine distinctions available in bibtex and map most everything onto the types of article, book, inbook, incollection, inproceedings, mastersthesis, phdthesis, and techreport. Preprints (a.k.a. "Manuscripts") are considered unnumbered techreports for our purposes, since they are often later distributed in that form. If present entries in the bibliography are any guide, you should rarely need other entry types. In particular, note that low-grade items like personal communications should not be included since our charter is to cover only openly available materials. If you need such an entry in your papers' reference lists, please keep it in a supplementary bibliography file until it is published. For example, "\bibliography{mine,mygroup,geom}" specifies a search path of three files bibtex can use to satisfy references. Citation tag: it's easy to come up with citetags that are mnemonic, short, or unique, but not to have all three at the same time. The system we use is a compromise. Our citetags consist of an author part (first letter of surname of each author), a title part (first letter or digit string of each significant word in the title, up to 7 characters), and a year part (last two digits of year of publication), separated by dashes. Thus "J. O'Rourke, Art Gallery Theorems and Algorithms, 1987" reduces to "o-agta-87". The tricky part of this is how to define "significant" words of the title, particularly when punctuation and mathematical strings are involved. Here are the formal rules, which are intended to produce a commonsense result as often as possible: - remove articles, conjunctions, and prepositions (i.e. words which wouldn't be capitalized in a title) - convert Roman numerals to Arabic, remove diacriticals and braces, remove quotes and apostrophes, convert other punctuation to spaces - retain only the first alpha/numeric token within $...$ delimiters - take the first letter, or first digit string, of remaining words - take the first 5 characters so produced For just under 99% of entries the citetag generated by this procedure will not conflict with that of any existing entry. But if it does, you'll have to find some way to break the tie. In our experience, collisions at this stage have come about only from various forms of the ``same paper'' conflicting with each other, and one of the following tiebreaking rules suffices: - for multipart papers, add the part number, in Arabic, to the title field (c-lbors1-90, c-lbors2-90) - for other variations on a theme, add a letter from a distinguishing word to the title field (s-mmdpsl-90, s-mmdpsp-90) - for alternate publications of a paper, append "a" for article, "i" for incollection or inbook or inproceedings, or "t" for techreport, to the year field (kkt-ptots-90i, kkt-ptots-90t) - otherwise, punt and discriminate using whatever you can (g-gramq-cga-88, g-gramq-edbt-88) If the resulting citetags don't match your favourite descriptor for the reference, you can still use the old familiar version if you declare a mapping between the two in your TeX source, such as the following: \newcommand{\smawk}{akmsw-gamsa-87} You needn't go through the process for entries you don't cite: the merging software will automatically generate citetags for contributed entries lacking them. It also keeps entries in the colt.bib file sorted in order of author, title, and year, in order to bring various appearances of the same paper together. (This should match the order of the default softcopy.) Fields: we support all fields from the bibtex standard styles, plus a few common extensions like abstract=, annote=, isbn=. Fields cites=, comments=, keywords=, precedes=, succeeds=, and update= are our own. Conventions for entering all of these are as follows. Quotes aren't necessary when the field value is entirely digits (true for volume, number, and year, normally). You should use the empty string "" for fields you can't complete just yet (e.g. pages = "" for a conference or journal paper to appear). Some older entries use a visible placeholder like "??"; if you need to cite them, please fill in the hole, or change to "", as feasible. abstract: verbatim from the original item (optional) - little used in present entries, it's best added only when short and sweet address: city of publication - use for books, techreports, theses, and obscure irregular conferences; otherwise discouraged - use only first city if publisher lists several - add two-letter state/province codes for US/Canada cities; for others, add country - give English-language name, with correct diacriticals (e.g. Munich rather than M{\"u}nchen, Saarbr{\"u}cken rather than Saarbruecken) - city/country names to use are those in effect at time of publication (e.g. West and East Germany from 1949 until 3 October 1990, Germany thereafter) annote: explanatory or critical comment on item content (optional) - little used in present entries, but welcomed author: - separate multiple author fields with " and ", order same as in reference - author's names in name, initials order - use braces to enclose capitalized or comma-separated elements of a compound surname, e.g. {Van Wyk} or {Lipski, Jr.} - instead of full given names you may follow the custom of mathematical literature and use initials, space-separated (exceptions to avoid collision: Ta. Asano, Te. Asano) - [van Leunen p.155] by "strict and narrow propriety" we should cite precisely the name which appears on the item, even if it leads to irregularities. While it is reasonable to fix up such typographical glitches (attributable to coauthors, copy editors, and the pressure of deadlines) as you are certain the author would want you to, inconsistent practice is for the author and not the bibliographer to worry about. booktitle: title of book or proceedings containing item - for English items, capitalize first word, first word after a colon, and all other words except articles and unstressed conjunctions and prepositions. Otherwise follow capitalization conventions of the native language, if you know them. (According to the MLA Handbook, for French, German, Italian, Latin, and Spanish, capitalization in titles is the same as in normal prose.) There is no need for braces on capitalized words in this field. - abbreviations for some popular conferences are in the authority file. The merging software will recognize and convert most variant abbreviations to standard form. chapter: chapter or section number, where item is part of a monograph - use entry type of incollection if chapter has its own title, inbook otherwise cites: citations made by item (optional) - give as list using biblio citetags, such as cites = "bs-dcms-76, gjpt-tsp-78, o-agta-87" - needn't be an exhaustive copy of the item's citations, but if used should at least give the significant ones. You can say cites = "(complete) bs-dcms-76, ..." if the list is exhaustive. comments: bibliographic marginal notes - supplemental information not a part of the reference proper: notes on a item's source language, or relation to other items, or a UMI order number and page count, or a Computing Reviews or Math Reviews number... - separate multiple comments with a semicolon - "to appear in", "submitted to", "in press" and the like require fixup later, at which time changes in other info such as title (and thus label) tend to be overlooked; so please use these only as comments on the future of an entry already published in some form edition: of a book - use numbered ordinal, e.g. "2nd" editor: - editors of proceedings not needed, and discouraged - otherwise, use guidelines for author institution: publisher of a techreport - include any relevant department, and list in minor-to-major order (e.g. "Dept. Comput. Sci., Univ. California Berkeley") isbn: of book (optional) - worthwhile only for obscure or otherwise hard-to-find items - give with hyphens as specified by publisher journal: - abbreviations for some popular journals are in the authority file. The merging software will recognize and convert most variant abbreviations to standard form. - separate journal series are considered separate journals, e.g. journal = "J. Combin. Theory Ser. A" rather than series or volume A keywords: - use to supplement, for searching or descriptive purposes, terms already present in the item's title - separate multiple keywords with commas - keywords need only be attached to the newest of a paper's appearances, if identical for all - use those in authority file, by preference - additions to the list of keywords, are expected and welcomed, within reason; month: month of publication - encouraged for techreports and theses, discouraged otherwise - use bibtex standard abbreviations (three letters, lower case, no quotes) note: - use for supplemental information which should appear in a citing paper's reference list; otherwise use comments field - e.g. note = "Errata in 2(1981), 105" - for theses, give techreport type and number, if known, e.g. note = "Report TR-86-103" number: of techreport, work in a series, or issue of journal - essential for true techreports (nolle techreportum sine numeratum) - for journals, necessary iff there exists more than one "page 1" per volume (e.g. proceedings as separately-paginated issue of journal), and discouraged otherwise - use "--" for combined issues, e.g. "3--4" pages: - use double dash "--" in a number range precedes/succeeds: pointer lists for temporal relationships among entries - for example precedes = "oy-nubks-88" points to new & improved paper succeeds = "k-cmgcl-77" is backpointer from it publisher: - see authority file for standard names of some publishers school: granting degree, for thesis - include any relevant department, since this assists inquiries about availability or contents, and list in minor-to-major order (e.g. "Dept. Comput. Sci., Univ. California Berkeley") series: of books - e.g. "Lecture Notes in Computer Science" title: of item - for English non-books, you need only capitalize first word and proper names, and enclose latter capitalized words in braces so that bibtex will leave them alone. If you prefer, you may capitalize other words in a title to get full uppers & lowers capitalization, but take care not to embrace them. (Full capitalization is optional because it's more complex, and we know of no journals still requiring it.) - omit qualifiers like "(extended abstract)" - [van Leunen p.170] regardless of the style of the original, use colon to separate title from subtitle (edit if necessary): for example change "Serial science. {I}. Definitions" to "Serial science, {I}: Definitions" - otherwise "correct" only what you're certain the author would want you to - enclose math expressions (including numbers) in {$ ... $}, and express in TeX notation type: of techreport or thesis - e.g. "Technical Report" or "Manuscript" or "M.{Phil}. Thesis" - for theses, give the actual degree name, and supplement with keywords "master thesis" or "doctoral thesis" accordingly - if the thesis was distributed as a numbered report, then give its type and number in the note= field - capitalized words after the first need braces in this field update: date and bibliographer corresponding to last change - maintained by the merging software (so don't bother) volume: in journal or multivolume book year: year of publication - "to appear", "submitted", "in press", etc? See the comments field. '%' lines before entries: marginal comments wrt bibliography maintenance - use to flag entries with errors you can't fix just now ("% wrong volume number"), or to flag truthful data that may look erroneous ("% yes, ``connexion''") - the string "###" can be used to call attention where you believe something is missing or wrong. Feel free to fix such entries if you have the correct details handy. MISCELLANEOUS COMMENTS and OPEN PROBLEMS Johnson's STOC/FOCS index issued in 1991 contains a wealth of information on papers originally published through those conferences. Look to it first if you need supplementary information on such a paper. Standard bibtex comes with a limit of 750 entries per bibliography, which is laughably small for us. People will be able to produce hardcopy according to their own tastes only once they get their local bibtex reconfigured. bibtex (or at least its standard style files) have some fixed ideas about how things can be published; for instance, the following won't print completely and elicits a warning on the grounds that a proceedings cannot have both volume and issue numbers (i.e. be published as a journal issue). However SIGGRAPH's position is that its papers are published *in* a proceedings and that the proceedings is *in* an issue of the journal. @inproceedings{awg-psg-78 , author = "P. Atherton and K. Weiler and D. P. Greenberg" , title = "Polygon shadow generation" , booktitle = "Proc. SIGGRAPH '78" , journal = "Comput. Graph." , volume = "12" , number = "3" , year = "1978" , pages = "275--281" , oldlabel = "geom-20" } Patashnik's Bibtexing notes suggest that we ``don't take the field names too seriously'' and adjust the entry until it prints right. For the moment, entries of this sort (the only ones are from SIGGRAPH) have been given the article type, and the booktitle field won't print. But if SIGGRAPH's position is correct, then it is bibtex which should adapt in the long run. (Perhaps bibtex might better have used a general inclusion field ``in = "siggraph78"'' rather than crossref with its hardwired list of permitted situations.) BIB SOFTWARE AVAILABLE ELSEWHERE The computational geometry community has made some programs available for working with bibliographies. They can be retrieved via anonymous ftp to cs.usask.ca [128.233.128.5], in file pub/geometry/geombib.tar.Z. They include programs to translate between `bibtex' format and the older `refer' format, and programs to search quickly in large bibliographies. A new program "bibview" which is an X tool for manipulating bibtex databases has been published in comp.sources.x/v18i099. Its README says it "supports the user in making new entries, searching for entries and moving entries from one bib to another. It is possible to work with more than one bib simultaneously. bibview is implemented with Xt and Athena Widgets." bibview is available for ftp from (among other places) comp.sources.x archive sites, from cs.orst.edu:/pub/src/printers/bibview.tar.Z, and from dsrbg2.informatik.tu-muenchen.de:/pub/tex/bibview-1.0.tar.Z. ------------------------------ From: Peter Turney Date: Tue, 19 Sep 1995 14:49:37 +0500 Subject: Second and Final CFP: special issue of Evolutionary Computation Second and Final Call for Papers EVOLUTIONARY COMPUTATION Special Issue on EVOLUTION, LEARNING, AND INSTINCT: 100 YEARS OF THE BALDWIN EFFECT In 1896, James Mark Baldwin proposed that individual learning can explain evolutionary phenomena that appear to require Lamarckian inheritance of acquired characteristics. The ability of individuals to learn can guide the evolutionary process. In effect, learning smoothes the fitness landscape, thus facilitating evolution. A special issue of Evolutionary Computation is planned for 1996, the 100th anniversary of Baldwin's paper. See "http://ai.iit.nrc.ca/baldwin/cfp.html" on the World Wide Web or send a message to peter@ai.iit.nrc.ca. Manuscripts due: February 1, 1996 Acceptance notification: May 1, 1996 Final manuscript due: August 1, 1996 Planned Publication date of issue: December 1996 Guest Editors: Peter D. Turney, National Research Council, Canada Darrell Whitley, Colorado State University, USA Russell W. Anderson, University of California, USA ------------------------------ From: Peter Turney Date: Wed, 20 Sep 1995 14:01:01 +0500 Subject: Workshop on Data Engineering for Inductive Learning On August 20, 1995, there was a workshop at IJCAI-95 on the topic of Data Engineering for Inductive Learning (DEIL). Some of the problems discussed in the DEIL workshop overlap with the concerns of the Machine Learning community. The papers presented at the workshop are now available on the web as compressed PostScript files: http://ai.iit.nrc.ca/DEIL/ The web site includes abstracts in HTML, papers in PostScript, the original CFP for the workshop, and pointers to related material. ------------------------------ From: "Richard K. Belew" Date: Wed, 20 Sep 1995 16:31:11 -0600 Subject: Foundations of GAs 1996 - CFP FOGA4 August 3-5, 1996 San Diego, California The 1996 Foundations of Genetic Algorithms (FOGA4) workshop will be the fourth biennial meeting of a workshop designed to explore theoretical issues relevant to genetic algorithms (GAs) and evolutionary compuation generally. FOGA4 will held Saturday August 3 through Monday, August 5, 199 at the Unviersity of San Diego in San Diego, California. A reception will be held Friday evening August 2, with regular workshop activities beginning August 3. Attendance at the workshop will be limited; the goal is to create a small forum with close interaction among all the participants. Individuals submitting papers will be given priority for attendance, and some slots will be reserved for students. All individuals interested in attending must indicate this by either submitting a paper or requesting attendance. Extended abstracts as well as requests for attendance must be received by February 1, 1996. Submissions should address theoretical issues in GAs or evolutionary computation. Papers which make use of empirical data should ensure that the experimental results are well-connected to foundational issues. Work describing the application of GAs to engineering problems is not appropriate for this meeting unless it clearly illuminates a larger theoretical concern. Extended abstracts should be no longer than 10 pages (11 point font). Electronic submission of problem-free Postscript files is encouraged. Four copies of hardcopy manuscripts must be postmarked no later than February 1, 1996. Authors of accepted papers will be notified by April 15, 1996. Drafts of the full paper are due by July 15, 1996 and will be distributed as part of a preprint to participants at the FOGA4 meeting. Authors of papers presented at the FOGA4 workshop will be asked to contribute final versions of their papers (based on discussion/feedback at the meeting) as part a volume to be published in book form. Important dates: 1 Feb 96 Extended abstract due 15 Apr 96 Authors notified of acceptance 15 Jul 96 Drafts of full paper due 3-5 Aug 96 FOGA4, San Diego, CA Paper submissions and inquires may be directed to: Richard K. Belew Computer Science & Engr. Dept. (0114) Univ. California -- San Diego La Jolla, CA 92093-0114 rik@cs.ucsd.edu Michael D. Vose C.S. Dept., 107 Ayres Hall The University of Tennessee Knoxville, TN 37996 vose@cs.utk.edu ------------------------------ Date: Fri, 15 Sep 95 16:40:21 +0100 From: A.Sharkey@dcs.shef.ac.uk Subject: SPECIAL ISSUE of Connection Science PRELIMINARY CALL FOR PAPERS: Deadline February 14th 1996 A special issue of Connection Science Papers are sought for this special issue of Connection Science. The aim of this special issue is to examine when, how, and why neural nets should be combined. The reliability of neural nets can be increased through the use of both redundant and modular nets, (either trained on the same task under differing conditions, or on different subcomponents of a task). Questions about the exploitation of redundancy and modularity in the combination of nets, or estimators, have both an engineering and a biological relevance, and include the following: * how best to combine the outputs of several nets. * quantification of the benefits of combining. * how best to create redundant nets that generalise differently (e.g. active learning methods) * how to effectively subdivide a task. * communication between neural net modules. * increasing the reliability of nets. * the use of neural nets for safety critical applications. Special issue editor: Amanda Sharkey (Sheffield, UK) Editorial Board: Leo Breiman (Berkeley, USA) Nathan Intrator (Brown, USA) Robert Jacobs (Rochester, USA) Michael Jordan (MIT, USA) Paul Munro (Pittsburgh, USA) Michael Perrone (IBM, USA) David Wolpert (Santa Fe Institute, USA) We solicit either theoretical or experimental papers on this topic. Questions and submissions concerning this special issue should be sent by February 14th 1996 to: Dr Amanda Sharkey, Department of Computer Science, Regent Court, Portobello Street, University of Sheffield, Sheffield, S1 4DP, United Kingdom. Net: amanda@dcs.shef.ac.uk ------------------------------ From: Robert Elliott Smith Date: Fri, 08 Sep 95 14:25:58 -0600 Subject: Special AI session of SECTAM XVIII AI Techniques in Engineering and Mechanics A Special Session of EIGHTEENTH SOUTHEASTERN CONFERENCE ON THEORETICAL AND APPLIED MECHANICS April 14 - 16, 1996 TUSCALOOSA, ALABAMA Artificial Intelligence (AI) techniques are now being used by the practicing engineer to solve a whole range of heretofore intractable problems. This session will consist of paper presentations describing the practical application of AI in all branches of engineering and mechanics, and thus will serve as a forum for the transfer of knowledge in this rapidly developing field. Papers are welcome from individuals and research groups on the subject of applications of AI including, but not limited to, the following topics: Systems and techniques such as: Expert systems, knowledge acquisition, knowledge-based systems, interactive knowledge-based systems, intelligent CAD/CAM systems, signal processing, sensor and data fusion, adaptive learning systems, neural networks, performance analysis, machine-vision systems, deductive databases, knowledge representation, modelling, learning heuristics, intelligent control systems, fuzzy logic, and genetic algorithms. Engineering applications including: Manufacturing, industrial engineering, production engineering, chemical engineering, civil engineering, electrical engineering, mechanical engineering, process control, robotics, autonomous vehicles, and communication. To submit a paper to the special session, follow instructions in the general SECTAM XVIII announcement (included below). Please indicate that the paper is for the AI special session. Contact persons for information on the special session are: Robert E. Smith Phone: (205) 348-1618 Email: rob@comec4.mh.ua.edu and Charles L. Karr Phone: (205) 348-0066 Email: ckarr@buster.eng.ua.edu Department of Engineering Science and Mechanics University of Alabama Box 870278 Tuscaloosa, AL 35487 Fax: (205) 348-7240 http://hamton.eng.ua.edu/college/orgs/esm/sectam.html ------------------------------ From: "Dr. Xu Lei" Date: Mon, 11 Sep 1995 16:46:37 +0800 Subject: ICONIP96 FIRST CALL FOR PAPERS 1996 INTERNATIONAL CONFERENCE ON NEURAL INFORMATION PROCESSING The Annual Conference of the Asian Pacific Neural Network Assembly ICONIP'96, September 24 - 27, 1996 Hong Kong Exhibition and Convention Center, Wan Chai, Hong Kong The goal of ICONIP'96 is to provide a forum for researchers and engineers from academia and industry to meet and to exchange ideas on the latest developments in neural information processing. The conference further serves to stimulate local and regional interests in neural information processing and its potential applications to industries indigenous to this region. CONFERENCE TOPICS ================= * Theory * Algorithms & Architectures * Applications * Supervised/Unsupervised Learning * Hardware Implementations * Hybrid Systems * Neurobiological Systems * Associative Memory * Visual & Speech Processing * Intelligent Control & Robotics * Cognitive Science & AI * Recurrent Net & Dynamics * Image Processing * Pattern Recognition * Computer Vision * Time Series Prediction * Financial Engineering * Optimization * Fuzzy Logic * Evolutionary Computing * Other Related Areas CONFERENCE'S SCHEDULE ===================== Submission of paper February 1, 1996 Notification of acceptance May 1, 1996 Early registration deadline July 1, 1996 SUBMISSION INFORMATION ====================== Authors are invited to submit one camera-ready original and five copies of the manuscript written in English on A4-format white paper with one inch margins on all four sides, in one column format, no more than six pages including figures and references, single-spaced, in Times-Roman or similar font of 10 points or larger, and printed on one side of the page only. Electronic or fax submission is not acceptable. Additional pages will be charged at USD $50 per page. Centered at the top of the first page should be the complete title, author(s), affiliation, mailing, and email addresses, followed by an abstract (no more than 150 words) and the text. Each submission should be accompanied by a cover letter indicating the contacting author, affiliation, mailing and email addresses, telephone and fax number, and preference of technical session(s) and format of presentation, either oral or poster (both are published). All submitted papers will be refereed by experts in the field based on quality, clarity, originality, and significance. Authors may also retrieve the ICONIP style, "iconip.tex" and "iconip.sty" files for the conference by anonymous FTP at ftp.cs.cuhk.hk in the directory /pub/iconip96. For further information, inquiries, and paper submissions please contact ICONIP'96 Secretariat Department of Computer Science The Chinese University of Hong Kong Shatin, N.T., Hong Kong Fax (852) 2603-5024 E-mail: iconip96@cs.cuhk.hk http://www.cs.cuhk.hk/iconip96 ------------------------------ End of ML-LIST (Digest format) **************************************** From mandynev@ece.rice.edu Fri Sep 22 11:51:54 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Fri, 22 Sep 95 11:51:40 -0500; AA03987 Received: from bilby.cs.uwa.oz.au by lucy.cs.wisc.edu; Fri, 22 Sep 95 11:51:36 -0500 Received: from (mafm@parma.cs.uwa.oz.au [130.95.1.7]) by cs.uwa.oz.au (8.6.8/8.5) with SMTP id SAA24703; Fri, 22 Sep 1995 18:08:15 +0800 Message-Id: <199509221008.SAA24703@cs.uwa.oz.au> From: Amanda Nevin Sender: mandynev@ece.rice.edu To: reinforce@cs.uwa.edu.au Subject: ICNN-96 June 2-6, 1996 Washington, DC Date: Thu, 21 Sep 95 14:43:52 -0500 (http://www-ece.rice.edu/96icnn/) INTERNATIONAL CONFERENCE ON NEURAL NETWORKS (ICNN'96) Sheraton Washington Hotel Washington, DC, USA TUTORIALS: June 2, 1996 CONFERENCE: June 3-6, 1996 Important Dates and Deadlines October 1, 1995 Proposals to Special Sessions Chair October 16, 1995 Papers received by Program Chair November 1, 1995 Proposals to Panel Sessions Chair December 1, 1995 Exhibit requirements to General Chair January 23, 1996 Authors informed of decisions February 23, 1996 Final papers to Program Chair General Chair Program Chair Benjamin Wah Bing Sheu Coordinated Science Lab. Dept. of Electrical Engr. Univ. of Illinois, Univ. of Southern California at Urbana-Champaign EP/Powell Hall 604 Urbana, IL 61801 Los Angeles, CA 90089-0271 icnn96@manip.crhc.uiuc.edu icnn96@pacific.usc.edu phone: (217) 333-3516 phone: (213) 740-4711 fax: (217) 244-7175 fax: (213) 740-8677 Panel Session Chair Special Session Chair Mohammed Ismail Jacek M. Zurada Dept. of Electrical Engr. Dept. of Electrical Engr. Ohio State University Univ. of Louisville 2015 Neil Avenue Louisville, KY 40292, USA Columbus, Ohio 43210-1272 jmzura02@starbase. ismail@ee.eng.ohio-state.edu spd.louisville.edu phone: (614) 292-0351 phone: (502) 852-6314 fax: (614) 292-7596 fax: (502) 852-6807 HOMEPAGE. Conference information is maintained by the Publicity Chair, Joseph R. Cavallaro, on the World Wide Web at http://www-ece.rice.edu/96icnn -- Mandy Nevin (713) 527-4025 Dept. of ECE, MS-366 Rice University Houston, TX 77251-1892 mandynev@ece.rice.edu From priel@eder.cc.biu.ac.il Fri Sep 22 15:18:05 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Fri, 22 Sep 95 15:17:56 -0500; AA07382 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Fri, 22 Sep 95 15:17:51 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id ab29637; 22 Sep 95 13:16:30 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id ab29634; 22 Sep 95 12:43:26 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa08173; 22 Sep 95 12:42:59 EDT Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id aa01698; 22 Sep 95 7:49:51 EDT Received: from eder.cc.biu.ac.il by EDRC.CMU.EDU id aa09697; 22 Sep 95 7:49:18 EDT Received: by eder.cc.biu.ac.il (AIX 3.2/UCB 5.64/4.03) id AA18392; Fri, 22 Sep 1995 13:48:32 +0200 From: priel@eder.cc.biu.ac.il Message-Id: <9509221148.AA18392@eder.cc.biu.ac.il> Subject: paper on analytical study of time-series generation ... To: connectionists@cs.cmu.edu Date: Fri, 22 Sep 1995 13:48:32 +0200 (WET) X-Mailer: ELM [version 2.4BIU PL20] Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1198 FTP-host: archive.cis.ohio-state.edu FTP-filename: /pub/neuroprose/kanter.time_series.ps.Z The file kanter.time_series.ps.Z is now available for copying from the Neuroprose repository. It has been accepted for publication in the PRL. (4 pages) Analytical Study of Time Series Generation by Feed-Forward Networks ---------------------------------------------- I. Kanter, D. A. Kessler, A. Priel and E. Eisenstein Minerva Center and Department of Physics, Bar-Ilan University, Ramat-Gan 52900, Israel ABSTRACT : Generation of time series is studied analytically for a generalization of the Bit-Generator to continuous activation functions and multi-layer architectures. The network exhibits at asymptotically large times the following characteristic features: (a) flows can be periodic or quasi-periodic depending on the phase of the weights, (b) the dimension of the attractor is a function of the gain of the activation function and the number of hidden units, (c) a phase shift in the weights results in a frequency shift in the output, so that the system operates as a phase detector. *** NO HARD COPIES ARE AVAILABLE *** From cnna96@cnm.us.es Fri Sep 22 15:40:02 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Fri, 22 Sep 95 15:40:00 -0500; AA07655 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Fri, 22 Sep 95 15:39:56 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa29637; 22 Sep 95 13:13:44 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa29634; 22 Sep 95 12:43:20 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa08168; 22 Sep 95 12:42:40 EDT Received: from CS.CMU.EDU by B.GP.CS.CMU.EDU id aa02245; 22 Sep 95 8:37:32 EDT Received: from cnm1.cnm.us.es by CS.CMU.EDU id aa14756; 22 Sep 95 6:34:57 EDT Received: from cnm10.cnm.us.es by cnm1.cnm.us.es (4.1/SMI-4.1) id AA18897; Fri, 22 Sep 95 12:39:04 +0200 Date: Fri, 22 Sep 95 12:39:04 +0200 From: 4th Workshop on CNN's and Applications Message-Id: <9509221039.AA18897@cnm1.cnm.us.es> To: Connectionists@cs.cmu.edu, cells@tce.ing.uniroma1.it Subject: WWW page on CNNA'96 A World Wide Web page has been established for up-to-date information on CNNA96 (see below). http://www.cica.es/~cnm/cnna96.html The preliminary call for papers (already published in this list) follows. ------------------------------------------------------------------------------ PRELIMINARY CALL FOR PAPERS 4th IEEE INTERNATIONAL WORKSHOP ON CELLULAR NEURAL NETWORKS AND APPLICATIONS (CNNA-96) June 24-26, 1996 (Jointly Organized with NDES-96) Escuela Superior de Ingenieros de Sevilla Centro Nacional de Microelectronica Sevilla, Spain ------------------------------------------------------------------------------ ORGANIZING COMMITTEE: Prof. J.L. Huertas (Chair) Prof. A. Rodriguez-Vazquez Prof. R. Dominguez-Castro SECRETARY: Dr. S. Espejo TECHNICAL PROGRAM: Prof. A. Rodriguez-Vazquez PROCEEDINGS: Prof. R. Dominguez-Castro SCIENTIFIC COMMITTEE: Prof. N.N. Aizemberg, Univ. of Uzhgorod, Ukrania Prof. L.O. Chua, Univ. of Cal. at Berkeley, U.S.A. Prof. V. Cimagalli, Univ. of Rome, Italy Prof. T.G. Clarkson, Kings College of London, U.K. Prof. A.S. Dmitriev, Academy of Sciences, Russia Prof. M. Hasler, EPFL, Switzerland Prof. J. Herault, Nat. Ins. of Tech., France Prof. J.L. Huertas, Nat. Microelectronics Center, Spain Prof. S. Jankowski, Tech. Univ. of Warsaw, Poland Prof. J. Nossek, Tech. Univ. Munich, Germany Prof. V. Porra, Tech. Univ. of Helsinki, Finland Prof. T. Roska, MTA-SZTAKI, Hungary Prof. M. Tanaka, Sophia Univ., Japan Prof. J. Vandewalle, Kath. Univ. Leuven, Belgium ------------------------------------------------------------------------------ GENERAL SCOPE OF THE WORKSHOP AND VENUE The CNNA series of workshops aims to provide a biannual international forum to present and discuss recent advances in Cellular Neural Networks. Following the successful conferences in Budapest (1990), Munich (1992), and Rome (1994), the fourth workshop will be held in Seville during 1996, organized by the National Microelectronic Center and the School of Engineering of Seville. Seville, the capital of Andalusia, and site of the 1992 Universal Exposition, combines a rich cultural heritage accumulated during its more than 2500 years history with modern infrastructures in a stable and sunny climate. It boasts a large, prestigious university, several high-technology research centers of the Spanish Council of Research, and many cultural attractions. It is linked to Madrid by high-speed train and has an international airport serving several daily direct international flights, as well as many connections to international flights via Madrid. ------------------------------------------------------------------------------ PAPERS SUBMISSION Papers on all aspects of Cellular Neural Networks are welcome. Topics of interest include, but are not limited to: - Basic Theory - Applications - Learning - Software Implementations and CNN Simulators - CNN Computers - CNN Chips - CNN System Development and Testing Prospective authors are invited to submit 4 pages summaries of their papers to the Conference Secretariat. Authors of accepted papers will be asked to deliver camera-ready versions of their full-papers for publication in an IEEE-sponsored Proceedings. ------------------------------------------------------------------------------ AUTHOR'S SCHEDULE Submission of summaries: ................ January 31, 1996 Notification of acceptance: ............. March 31, 1996 Submission of camera-ready papers: ...... May 15, 1996 ------------------------------------------------------------------------------ PRELIMINARY REGISTRATION FORM Fourth IEEE Int. Workshop on Cellular Neural Networks and their Applications CNNA'96 Sevilla, Spain, June 24-26, 1996 I wish to attend the workshop. Please send Program and registration form when available. Name: ................______________________________ Mailing address: .....______________________________ Phone: ...............______________________________ Fax: ............. ...______________________________ E-mail: ..............______________________________ Please complete and return to: CNNA'96 Secretariat. Department of Analog Circuit Design, Centro Nacional de Microelectronica Edif. CICA, Avda. Reina Mercedes s/n, E-41012 Sevilla - SPAIN FAX: +34-5-4231832 Phone: +34-5-4239923 E-mail: cnna96@cnm.us.es ------------------------------------------------------------------------------ From cia@kamo.riken.go.jp Sat Sep 23 06:17:21 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Sat, 23 Sep 95 06:17:18 -0500; AA14597 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Sat, 23 Sep 95 06:17:16 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa00820; 23 Sep 95 5:00:58 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa00818; 23 Sep 95 4:51:46 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa08943; 23 Sep 95 4:51:09 EDT Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id aa15696; 23 Sep 95 1:00:42 EDT Received: from kamo.riken.go.jp by EDRC.CMU.EDU id aa13319; 23 Sep 95 0:59:50 EDT Received: by kamo.riken.go.jp; id AA16627; Sat, 23 Sep 1995 14:04:14 +0900 Message-Id: <9509230504.AA16627@kamo.riken.go.jp> To: Connectionists@cs.cmu.edu Cc: cia@kamo.riken.go.jp Subject: Nolta-95 - Call for participation Date: Sat, 23 Sep 95 14:04:14 +0900 From: cia@kamo.riken.go.jp X-Mts: smtp Many reserchers aked me to send more information about Nolta 95 in Las Vegas. Enclosed, please find "Call for participation". Andrzej Cichocki Laboratory for Artificial Brain Systems, Frontier Research Program RIKEN, Institute of Physical and Chemical Research, Hirosawa 2-1, Saitama 351-01, WAKO-Schi, JAPAN E-mail: cia@kamo.riken.go.jp, FAX (+81) 048 462 4633. URL: http://zoo.riken.go.jp/bip.html ----------------------------------------------------------------------- CALL FOR PARTICIPATION 1995 International Symposium on Nonlinear Theory and its Applications (NOLTA'95) Caesars Palace, Las Vegas, Nevada, U.S.A. December 10 - 14, 1995 ----------------------------------------------------------------------- The 1995 International Symposium on Nonlinear Theory and its Applications (NOLTA'95) will be held at Caesars Palace, Las Vegas, Nevada, U.S.A. on Dec.10-14, 1995. The conference is open to all the world. About 300 papers describing original work in all aspects of Nonlinear Theory and its Applications are presented: [Plenary Talk] 1. CHUA, L.O. (U. C. Berkeley) : Nonlinear Waves, Patterns and Spatio-Temporal Chaos 2. HASEGAWA, Akira (Osaka Univ.): Recent Progress of Optical Soliton Research 3. WILLSON, Jr., A.N. (U.C.L.A.): Some Aspects of Nonlinear Transistor Circuit Theory [Special(invited) Sessions] 1. Application of Nonlinear Analysis to Communication 2. Blind Separation of Sources - Brain Information Processing - 3. Coupled Chaotic Systems - Modeling Brain Functions and Information Processing 4. Fundamental Advances in the Theory of Networks and Systems 5. Hardware Implementation of Nonlinear Dynamical Systems and Its Applications 6. Homotopy Continuation for Nonlinear Analysis 7. Information Processings and Complexity 8. Nonlinear Dynamics and Neural Coding 9. The CNN Nonlinear Dynamic Visual Microprocessor - New Chips, Applications, and Biological Relevance - [Regular Sessions] 1. Bifurcation 2. Biocybernetics and Evolution Systems 3. Cellular Neural Networks 4. Circuit Simulation and Modeling 5. Chaos and Spatial Chaos 6. Chaos Control 7. Chaotic Series Analysis 8. Chaotic Systems 9. Chua's Circuits 10.Communication 11.Electronic Circuits 12.Fractals 13.Fuzzy 14.Image and Signal Processing 15.Information Dynamics 16.Neural Networks (Applications I) 17.Neural Networks (Applications II) 18.Neural Networks (Applications III) 19.Neural Networks (Artificial Systems) 20.Neural Networks (Learning and Capacity) 21.Nonlinear Partial Differential Equations 22.Nonlinear Physics 23.Numerical Analysis and Validation 24.Numerical Methods and Nonlinear Circuits 25.Oscillation 26.Real Nonlinear Control 27.Synchronization 28.Theory of Nonlinear Control 29.Time Series Analysis and Economics 30.Transmission Line ----------------------------------------------------------------------- Registration Before Oct. 31, the registration fee is Japanese Yen 40,000(US$400.00) for researcher, and Yen 20,000(US$200.00) for full-time student. After Nov. 1, the fee is Yen 50,000(US$500.00) for researcher and Yen 25,000(US$250.00) for student. The registration fee includes attendance at all sessions, tea & coffee breaks, a copy of the symposium proceedings, and a banquet ticket. ----------------------------------------------------------------------- Accomodation By reserving through NOLTA'95 Reservation Cards, guest rooms of Caesars Palace during NOLTA'95 will be provided at the rate of US$109.00 for single or double occupancy. For further information, please contact NOLTA'95 secretariat (nolta@sat.t.u-tokyo.ac.jp). ----------------------------------------------------------------------- If you can use WWW browser, you can find further information at http://hawk.ise.chuo-u.ac.jp/NOLTA . If not, please contact to NOLTA'95 secretariat: NOLTA'95 secretariat c/o Oishi Lab., Dept. of Information and Computer Sciences School of Science and Engineering, Waseda University 3-4-1, Okubo, Shinjuku-ku, Tokyo 169, JAPAN Telefax : +81-3-5272-5742 e-mail : nolta@sat.t.u-tokyo.ac.jp ----------------------------------------------------------------------- Organizer : Research Society of Nonlinear Theory and its Applications, IEICE In cooperation with : IEEE Neural Networks Council International Neural Network Society Asian Pacific Neural Network Assembly IEEE CAS Technical Committee on Nonlinear Circuits and Systems Technical Group of Nonlinear Problems, IEICE Technical Committee of Electronic Circuits, IEEJ ----------------------------------------------------------------------- NOLTA'95 Symposium Committee HONORARY CHAIRS: Kazuo Horiuchi(Waseda Univ.) Masao Iri(Chuo Univ.) CO-CHAIRS: Shun-ichi Amari(Univ. of Tokyo) Shinsaku Mori(Keio Univ.) Allan N. Willson, Jr.(U.C.L.A.) TECHNICAL PROGRAM CHAIR: Shun-ichi Amari(Univ. of Tokyo) LOCAL ARRANGEMENT CHAIR: Allan N. Willson, Jr. Dept. of Electrical. Engr., University of Cal., Los Angeles CA 90024, U.S.A. Telefax : +1-310-206-4061 e-mail : willson@epsilon.icsl.ucla.edu PUBLICITY CHIR: Shinsaku Mori(Keio Univ.) PUBLICATION: Toshimichi Saito(Hosei Univ.) SECRETARIES: Shin'ichi Oishi (Waseda Univ.) oishi@oishi.info.waseda.ac.jp Toshimichi Saito(Hosei Univ.) saito@toshi.ee.hosei.ac.jp Mitsunori Makino(Chuo Univ.) makino@ise.chuo-u.ac.jp ----------------------------------------------------------------------- From cia@kamo.riken.go.jp Sat Sep 23 09:02:41 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Sat, 23 Sep 95 09:02:39 -0500; AA16237 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Sat, 23 Sep 95 09:02:37 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id ab00820; 23 Sep 95 5:01:57 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id ab00818; 23 Sep 95 4:51:47 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa08948; 23 Sep 95 4:51:25 EDT Received: from CS.CMU.EDU by B.GP.CS.CMU.EDU id aa16950; 23 Sep 95 3:59:28 EDT Received: from kamo.riken.go.jp by CS.CMU.EDU id aa24568; 23 Sep 95 3:58:39 EDT Received: by kamo.riken.go.jp; id AA16557; Sat, 23 Sep 1995 17:03:23 +0900 Message-Id: <9509230803.AA16557@kamo.riken.go.jp> To: Connectionists@cs.cmu.edu Cc: cia@kamo.riken.go.jp Subject: Postdoctoral positions available in JAPAN Date: Sat, 23 Sep 95 17:03:23 +0900 From: cia@kamo.riken.go.jp X-Mts: smtp Postdoctoral research positions are available to participate in Neural Information Processing- Frontier Research Program RIKEN. Applications are invited for postdoctoral positions to study the artificial neural systems in the laboratory for ARTIFICIAL BRAIN SYSTEMS in the RIKEN (Institute of Physical and Chemical Research -Wako -city, Japan). We are seeking a highly-motivated postdoctoral scientist working in the area of neural networks with a strong background in nonlinear and adaptive signal processing, and/or statistical computation, speech and image processing , mathematics and computer science. Salaries range between 4-6 million Yen (40k - 60k US $) per annum depending on experience, achievements and the number of years since the PhD was earned. The Institute of Physical and Chemical Research (RIKEN) have started a new eight-years Frontier Research Program on Neural Information Processing, beginning in October 1994. The Program includes three research laboratories ( Neural Modeling, Neural Information Representations and Artificial Brain Systems) , each consisting of one research leader and several researchers. We will study fundamental principles underlying the higher order brain functioning from mathematical, information-theoretic and systems-theoretic points of view. The three laboratories cooperate in constructing various models of the brain, mathematically analyzing information principles in the brain, and designing artificial neural systems. We will have close correspondences with another Frontier Research Program on experimental Neuroscience. Research positions, available from November1995- April 1996, or latter are opened for one-year contracts to researchers and post-doctoral fellows, and are renewable for next years depending on success in the program and expected continuation of funding. The search will be continued untill the position is filled. Applicants must submit a letter of application, a current full curriculum vitae/resume including list of publications, the name of three referees and a detailed statement of research interests by e-mail to: Dr. Andrzej Cichocki Team leader of Laboratory for ARTIFICIAL BRAIN SYSTEMS Frontier Research Program RIKEN, Institute of Physical and Chemical Research, Hirosawa 2-1, Saitama 351-01, WAKO-Schi, JAPAN E-mail: cia@kamo.riken.go.jp, FAX (+81) 048 462 4633. URL: http://zoo.riken.go.jp/bip.html The Riken (Institute of Physical and Chemical Research) is a research Institute supported by the Japanese Government. The number of tenured researchers in the Riken is about 350 and more than 1000 are working altogether. The Riken is located at Wako City, Saitama Prefecture, in the suburb of Tokyo. It is very close to central Tokyo. It takes only 15 minutes by train from Ikebukuro Station (one of the famous downtowns in Tokyo). The Frontier Research Program is basic-science oriented, and the goal of the computational Neuroscience group is to elucidate fundamental principles of information processing in the brain by theoretical approaches. The group consists of three teams. Each researcher may propose one's research themes through discussions with a team leader and he should take part in one or two more research projects proposed by a team leader. Official language in the research group is English. There are more than 100 foreigners working in Riken. *************************** Research program of Laboratory for Artificial Brain Systems 1. Objectives The brain realizes intelligent functions over its complex neural network systems. In order to understand and physically simulate the mechanisms of such complex systems, one important method is to synthesize and design artificial neural systems to see how they work under various environmental conditions. We hope that such approach not only make possible to better understand the principles and mechanisms of the brain but also opens new perspectives to develop neurocomputers and their applications. The main objective is development and investigation of models, architectures (structures) and associated learning algorithms of artificial neural systems. The main emphasis will be given to development of novel learning algorithms that are biologically justified and are computational efficient (e.g., they provide numerical stability and high convergence speed). The second objective is investigation of some potential and perspective applications of artificial neural systems in : information and signal processing, high speed parallel computing, solving in real time some optimization problems, classification and pattern recognition problems (e.g. face recognition). 2. Main Topics (1) Blind deconvolution and separation of sources (Cocktail party problem). (2) Development and investigation of on-line adaptive unsupervised learning algorithms. (3) Investigation of recurrent dynamic neural networks with controlling chaos. (4) Investigation of artificial neural systems with variable architectures. (6) Application of artificial neural systems for high speed parallel computing, recognition and optimization problems. 3. Main Features (1) This research is intended to study the system -theoretic frameworks for artificial neural networks based on nonlinear adaptive signal processing and dynamic systems/control theory. (2) The techniques and algorithms developed in this team will be available for collaboration with other teams and individual researchers (3) High priority will be given to the development of synthetic neural systems which are biologically plausible but also implementable (realizable) by electronic and/or optical circuits and systems. (4) Although we intend to investigate mainly models that have possibly biological plausibility (resemblance) and main inspiration will be taken from neuroscience, we assume that investigated structures and algorithms for specific applications may be rather loosely associated with real biological nervous models or they will be only simplified systems of such models. From ro2m@crab.psy.cmu.edu Sat Sep 23 22:41:57 1995 Received: from lucy.cs.wisc.edu by sea.cs.wisc.edu; Sat, 23 Sep 95 22:41:55 -0500; AA21767 Received: from TELNET-1.SRV.CS.CMU.EDU by lucy.cs.wisc.edu; Sat, 23 Sep 95 22:41:53 -0500 Received: from TELNET-1.SRV.CS.CMU.EDU by telnet-1.srv.cs.CMU.EDU id aa01868; 23 Sep 95 22:39:12 EDT Received: from DST.BOLTZ.CS.CMU.EDU by TELNET-1.SRV.CS.CMU.EDU id aa01861; 23 Sep 95 22:21:09 EDT Received: from DST.BOLTZ.CS.CMU.EDU by DST.BOLTZ.CS.CMU.EDU id aa09411; 23 Sep 95 22:20:42 EDT Received: from CS.CMU.EDU by B.GP.CS.CMU.EDU id aa25009; 23 Sep 95 14:51:51 EDT Received: from ANDREW.CMU.EDU by CS.CMU.EDU id aa26593; 23 Sep 95 14:51:31 EDT Received: from crab.psy.cmu.edu.psy.cmu.edu (CRAB.PSY.CMU.EDU [128.2.248.139]) by andrew.cmu.edu (8.6.12/8.6.12) with SMTP id OAA09745 for ; Sat, 23 Sep 1995 14:51:29 -0400 Received: by crab.psy.cmu.edu.psy.cmu.edu (4.1/SMI-4.1) id AA01665; Sat, 23 Sep 95 14:51:28 EDT Date: Sat, 23 Sep 95 14:51:28 EDT From: "Randall C. O'Reilly" Message-Id: <9509231851.AA01665@crab.psy.cmu.edu.psy.cmu.edu> To: connectionists@cs.cmu.edu Subject: ANNOUNCING: PDP++ version 1.0 ANNOUNCING: The PDP++ Software Authors: Randall C. O'Reilly, Chadley K. Dawson, and James L. McClelland The PDP++ software is a new neural-network simulation system written in C++. It represents the next generation of the PDP software released with the McClelland and Rumelhart "Explorations in Parallel Distributed Processing Handbook", MIT Press, 1987. It is easy enough for novice users, but very powerful and flexible for research use. The current version is 1.0, our first non-beta release. It has been extensively tested and should be completely usable. The software can be obtained by anonymous ftp from: Anonymous FTP Site: hydra.psy.cmu.edu/pub/pdp++/ *or* unix.hensa.ac.uk/mirrors/pdp++/ For more information, see our web page: WWW Page: http://www.cs.cmu.edu/Web/Groups/CNBC/PDP++/PDP++.html There is a 250 page (printed) manual and an HTML version available on-line at the above address. New Features Since Previous Release (1.0b): =========================================== o Better support for sub-groups of units within a Layer (including better interface support). o Fixed and much more flexible 'ReadOldPDP' function for importing environments (pattern files). o Improved documentation for compiling the software. o Pre-compiled InterViews libraries for g++ now available (in addtion to cfront-based ones). o Added a bpso++ executable, which allows creation of mixed backprop and self-organizing networks. o Lots of bug fixes (see the ChangeLog file for details). Software Features: ================== o Full Graphical User Interface (GUI) based on the InterViews toolkit. Allows user-selected "look and feel". o Network Viewer shows network architecture and processing in real- time, allows network to be constructed with simple point-and-click actions. o Training and testing data can be graphed on-line and network state can be displayed over time numerically or using a wide range of color or size-based graphical representations. o Environment Viewer shows training patterns using color or size-based graphical representations. o Flexible object-oriented design allows mix-and-match simulation construction and easy extension by deriving new object types from existing ones. o Built-in 'CSS' scripting language uses C++ syntax, allows full access to simulation object data and functions. Transition between script code and compiled code is simplified since both are C++. Script has command-line completion, source-level debugger, and provides standard C/C++ library functions and objects. o Scripts can control processing, generate training and testing patterns, automate routine tasks, etc. o Scripts can be generated from GUI actions, and the user can create GUI interfaces from script objects to extend and customize the simulation environment. Supported Algorithms: ===================== o Feedforward and recurrent error backpropagation. Recurrent BP includes continuous, real-time models, and Almeida-Pineda. o Constraint satisfaction algorithms and associated learning algorithms including Boltzmann Machine, Hopfield models, mean-field networks (DBM), Interactive Activation and Competition (IAC), and continuous stochastic networks. o Self-organizing learning including Competitive Learning, Soft Competitive Learning, simple Hebbian, and Self-organizing Maps ("Kohonen Nets"). The Fine Print: =============== PDP++ is copyrighted and cannot be sold or distributed by anyone other than the copyright holders. However, the full source code is freely available, and the user is granted full permission to modify, copy, and use it. See our web page for details. The software runs on Unix workstations under XWindows. It requires a minimum of 16 Meg of RAM, and 32 Meg is preferable. It has been developed and tested on Sun Sparc's under SunOs 4.1.3, HP 7xx under HP-UX 9.x, and SGI Irix 5.3. Statically linked binaries are available for these machines. Other machine types will require compiling from the source. Cfront 3.x and g++ 2.6.3 are supported C++ compilers (we specifically were *not* able to compile successfully with the 2.7.0 version of g++, and gave up waiting for 2.7.1). The GUI in PDP++ is based on the InterViews toolkit, version 3.2a. However, we had to patch it to get it to work. We distribute pre-compiled libraries containing these patches for the above architectures. For architectures other than those above, you will have to apply our patches to InterViews before compiling. The basic GUI and script technology in PDP++ is based on a type-scanning system called TypeAccess which interfaces with the CSS script language to provide a virtually automatic interface mechanism. While these were developed for PDP++, they can easily be used for any kind of application, and CSS is available as a stand-alone executable for use like Perl or TCL. The binary-only distribution requires about 54 Meg of disk space, since we have been unable to get shared libraries to work with C++ on the above platforms. Each simulation executable is around 8-12 Meg in size, and there are 4 of these (bp++, cs++, so++, and bpso++), plus the CSS and 'maketa' executables. The compiled source-code distribution takes about 115 Meg (but only around 16 Meg before compiling). For more information on the details of the software, see our web page. - Randy