Author: xavier

List of papers to be presented at IWLCS 2006

This is the list of papers accepted for presentation at IWLCS 2006 that will take place during GECCO 2006.

  • Empirical evaluation of ensemble techniques for a Pittsburgh Learning Classifier System
    Bacardit, J. and Krasnogor, N.
  • An Artificial Life Classifier System for Real-Valued Inputs
    Bishop, J.
  • Technology Extraction for Future Generations from Process Time Series Data Reflecting Expert Operator Skills
    Kurahashi, S. and Terano, T.
  • An Initial Analysis of Parameter Sensitivity for XCS with Computed Prediction
    Lanzi, P.L., and Zanini, M.
  • The χ-ary Extended Compact Classifier System: Linkage Learning in Pittsburgh LCS
    Llorà, X., Sastry, K., Goldberg, D.E., and delaOssa, L.
  • Using XCS for Action Selection in RoboCup Rescue Simulation League
    Martínez, I. C., Ojeda, D., and Zamora, E.
  • Extending XCS with Representation in First-Order Logic
    Mellor, D.
  • A Further Look at UCS Classifier System
    Orriols-Puig, A., Bernad&oaccute;-Mansilla, E.
  • Agent-Based Learning Classifier Systems for Grid Data Mining
    Santos, M.F, Quintela, H., and Neves, J.
  • Community of Practice under Learning Classifier Systems
    Suematsu, Y.I.L., Takadama, K., Shimohara, K., and Katai, O.
  • Developing Conversational Interfaces with XCS
    Toney, D., Moore, J., and Lemon, O.
  • Dual-structured Classifier System Mediating XCS and Gradient Descent based Update
    Wada, A., Takadama, K., and Shimohara, K.

Advances at the frontier of LCS: First step completed

The first step of the volume Advances at the frontier of LCS is almost done. Below there is a list of the camera readies collected so far. These book chapters cover the contributions to IWLCS on 2003 and 2004.

  • Data Mining in Learning Classifier Systems: Comparing XCS with GAssist.
    Bacardit, J. and Butz, M.
  • Bloat Control and Generalization Pressure using the Minimum Description Length Principle for a Pittsburgh approach Learning Classifier System.
    Bacardit, J. and Garrell, J.M.
  • Improving the Performance of a Pittsburgh Learning Classifier System Using a Default Rule.
    Bacardit, J., Goldberg, D.E., and Butz, M.
  • Effect of Pure Error-Based Fitness in XCS.
    Butz, M., Goldberg, D.E., and Lanzi, P.L.
  • A Formal Relationship Between Ant Colony Optimizers and Classifier Systems.
    Davis, D.
  • An Experimental Comparison between ATNoSFERES and ACS.
    Landau, S., Sigaud, O., Picault, S., and Gérard, P.
  • Where to Go Once You Have Evolved a Bunch of Promising Hypotheses?.
    Llorà, X., Bernadó, B., Bacardit, J., and Traus, I.
  • A Hyper-Heuristic Framework with XCS: Learning to Create Novel Problem-Solving Algorithms Constructed from Simpler Algorithmic Ingredients.
    Martín-Blazquez, J. and Schulenburg, S.
  • Backpropagation in Accuracy-based Neural Learning Classifier Systems .
    O’Hara, T. and Bull, L.
  • Use of Learning Classifier System for Inferring Natural Language Grammar .
    Unold, O. and Dabrowski, G.
  • Analyzing Parameter Sensitivity and Classifier Representations for Real-valued XCS .
    Wada, A., Takadama, K., Shimohara, K., and Katai, O.
  • Three Architectures for Continuos Action.
    Wilson, S.W.
  • Using XCS to Describe Continuous-Valued Problem Spaces.
    Wyatt, D., Bull, L., and Parmee, I.

Multi-Objective Machine Learning

The book Multi-objective Machine Learning edited by Yaochu Jin contains several chapters on the usage of LCS and GBML for multi-objective machine learning. Among other topics it includes the usage of multi-objective optimization to evolve accurate and compact rule sets using LCS and GBML, and the use of GA-based Pareto optimization for rule extraction from neural networks.

Rule-Based Evolutionary Online Learning Systems

This book by Martin Butz offers a comprehensive introduction to learning classifier systems (LCS) – or more generally, rule-based evolutionary online learning systems. LCSs learn interactively – much like a neural network – but with an increased adaptivity and flexibility. This book provides the necessary background knowledge on problem types, genetic algorithms, and reinforcement learning as well as a principled, modular analysis approach to understand, analyze, and design LCSs. The analysis is exemplarily carried through on the XCS classifier system – the currently most prominent system in LCS research. Several enhancements are introduced to XCS and evaluated. An application suite is provided including classification, reinforcement learning and data-mining problems. Reconsidering John Holland’s original vision, the book finally discusses the current potentials of LCSs for successful applications in cognitive science and related areas.

Ninth International Workshop on Learning Classifier Systems (IWLCS 2006) – CFP

Seattle, WA, USA, July 8-9, 2006. To be held during the Genetic and Evolutionary Computation Conference (GECCO-2006), July 8-12, 2006.

Since Learning Classifier Systems (LCSs) were introduced by Holland as a way of applying evolutionary computation to machine learning problems, the LCS paradigm has broadened greatly into a framework encompassing many representations, rule discovery mechanisms, and credit assignment schemes. Current LCS applications range from data mining to automated innovation to on-line control. Classifier systems are a very active area of research, with newer approaches, in particular Wilson’s accuracy-based XCS, receiving a great deal of attention. LCS are also benefiting from advances in the field of reinforcement learning, and there is a trend toward developing connections between the two areas.

We invite submissions which discuss recent developments in all areas of research on, and applications of, Learning Classifier Systems.

IWLCS is the only event to bring together most of the core researchers in classifier systems. A free introductory tutorial on LCS will be presented at GECCO 2006.

Submissions

There are two possibilities for paper submissions. Both will be peer reviewed, but reviews of short papers will be mainly to provide feedback to authors – we expect most or all will be accepted.

1) Short papers of up to 4 pages may be submitted. Accepted short papers will be presented at the workshop and published in the GECCO workshop volume. The format of the GECCO workshop volume is to be confirmed but we expect it will be the ACM format used in 2005. After the workshop authors will be invited to submit full papers which are reviewed again for the post-workshop proceedings, which we plan to publish in Springer’s LNAI series as in past years.

2) Full papers of up to 20 pages (in Springer format) may be submitted for peer review before the workshop. Accepted full papers will be presented at the workshop and will be published in the post-workshop proceedings. Authors of full papers have a choice of how to contribute to the GECCO workshop volume: either i) prepare a short version for GECCO or ii) publish only your abstract in the GECCO book. If you prefer i) we would suggest an extended abstract of 1 or 2 pages, but anything up to 50% of the full paper is ok.

Papers should be submitted as PDF files e-mailed to iwlcs@cas.dis.titech.ac.jp.

Important dates

Please note: all dates are to be confirmed.

  • Paper submission deadline: March 24, 2006 (extended)
  • Decisions: April 12, 2006 (extended)
  • GECCO 2006 Workshop proceedings camerar-ready: April 26, 2006 (extended)
  • Workshop: July 8-9, 2006

Camera Ready for GECCO 2006 Workshop Proceedings

The camera-ready papers should be formated following the instructions provided by GECCO. Failing to comply will result in exclusion from the proceedings. The proceedings will only be published on CD-ROM. Camera-ready papers must be submitted using the GECCO-2006 Submission & Review site at https://ssl.linklings.net/conferences/gecco2006/.

Organization

Organizing Commitee

  • Tim Kovacs, University of Bristol (UK)
  • Xavier Llorà, University of Illinois at Urbana-Champaign (USA)
  • Keiki Takadama, Tokyo Institute of Technology (Japan)

Advisory Committee

  • Pier Luca Lanzi, Politechnico de Milano (Italy)
  • Wolfgang Stolzmann, Daimler Chrysler AG (Germany)
  • Stewart Wilson, Prediction Dynamics (USA)

Program Committee

  • Bacardit, Jaume. University of Nottingham (UK)
  • Bagnall, Tony. Univesity of East Anglia (UK)
  • Barry, Alwyn. University of Bath (UK)
  • Bernadó Mansilla, Ester. Universitat Ramon Llull (Spain)
  • Bonarini, Andrea. Politecnico di Milano (Italy)
  • Booker, Lashon. The Mitre Corporation (USA)
  • Browne, Will. University of Reading (UK)
  • Bull, Larry. University of West England (UK)
  • Butz, Martin. Universitat Wurzburg (Germany)
  • Carse, Brian. University of West England (UK)
  • Davis, David. NuTech Solutions (USA)
  • Drugowitsch, Jan. University of Bath (UK)
  • Egginton, RobUniversity of Bristol (UK)
  • Herrera, Francisco. Universidad de Granada (Spain)
  • Holmes, John. University of Pennsylvania (USA)
  • Homaifar, Abdollah. North Carolina A&T State University (USA)
  • Kovacs, Tim. University of Bristol (UK)
  • Lanzi, Pier Luca. Politecnico di Milano (Italy)
  • Llorà, Xavier. University of Illlinois at Urbana-Champaign (USA)
  • Marin-Blazquez, Javier. Universidad de Murcia (Spain)
  • Miramontes-Hercog, Luis. Instituto Tecnológico y de Estudios Superiores de Monterrey (Mexico)
  • Muruzabal, Jorge. Universidad Rey Juan Carlos (Spain)
  • Schulenburg, Sonia. University of Edinburgh (UK)
  • Sigaud, Olivier. Laboratoire d’Informatique de Paris 6 (France)
  • Stolzman, Wolfgang. Daimler Chrysler AG (Germany)
  • Takadama, Keiki. Tokyo Institute of Technology (Japan)
  • Wada, Atsushi. Advanced Telecomunications Research Institute (Japan)
  • Wilson, Stewart. Prediction Dynamics (USA)
  • Zatuchna, Z. V. Univesity of East Anglia (UK)

For further information please contact iwlcs@cas.dis.titech.ac.jp.

Advances at the frontier of LCS (Volume I) is coming

The final editing of the volume Advances at the frontier of LCS to be published by Springer is advancing at steady pace. The volume is going to be an overview of the research LCS and other GBML presented at IWLCS. The volume will cover 2003, 2004, and 2005 contributions.

So far, these are the raw numbers for 2003 and 2004 contributions:

  • 2003: 11 chapters by 26 different authors
  • 2004: 8 chapters by 15 different authors

The decisions about 2005 will be out soon. We will keep you posted

Camera ready instructions for IWLCS 2003 and 2004 proceedings

Springer has agreed to publish the compilation volume Advances in Learning Classifier Systems (the title may be slightly changed) including contributions from the International Workshop of Learning Classifier Systems in its editions of 2003, 2004, and 2005. This volume will present an overview of the work presented in the last three years of the workshop and will include up to 30 contributions.

The deadline for the camera-ready of your contribution to IWLCS was initially set to November 15. Due to the previous delay, we would extend this deadline until November 25 for your convenience. Please do not to hesitate to get in touch if you may not be able to reach this deadline. Due to the size of this volume, we would like to stick to this deadline to be able to have the volume ready for the next workshop edition in Seattle.

For further instructions about how to prepare your camera ready please check the Springer format instructions for authors at

Contributions should not exceed 20 pages. Authors providing camera- readies that do not complain with the LNCS format or exceed the maximum number of pages will be ask to resubmit them, and may not be included if time constraints do not allow us to do so.