Design and Analysis of Learning Classifier Systems: A Probabilistic Approach

The book Design and Analysis of Learning Classifier Systems: A Probabilistic Approach by Jan Drugowitsch presents a machine learning approach to Learning Classifier Systems. In the author’s own words:

This book provides a comprehensive introduction to the design and analysis of Learning Classifier Systems (LCS) from the perspective of machine learning. LCS are a family of methods for handling unsupervised learning, supervised learning and sequential decision tasks by decomposing larger problem spaces into easy-to-handle subproblems. Contrary to commonly approaching their design and analysis from the viewpoint of evolutionary computation, this book instead promotes a probabilistic model-based approach, based on their defining question “What is an LCS supposed to learn?”. Systematically following this approach, it is shown how generic machine learning methods can be applied to design LCS algorithms from the first principles of their underlying probabilistic model, which is in this book  for illustrative purposes  closely related to the currently prominent XCS classifier system. The approach is holistic in the sense that the uniform goal-driven design metaphor essentially covers all aspects of LCS and puts them on a solid foundation, in addition to enabling the transfer of the theoretical foundation of the various applied machine learning methods onto LCS. Thus, it does not only advance the analysis of existing LCS but also puts forward the design of new LCS within that same framework.

IWLCS 2008 call for papers

The Eleventh International Workshop on Learning Classifier Systems (IWLCS 2008) will be held in Atlanta, Georgia, USA, Sunday, July 13, 2008 during the Genetic and Evolutionary Computation Conference (GECCO-2008), July 12-16, 2008.

Originally, Learning Classifier Systems (LCSs) were introduced by John H. Holland as a way of applying evolutionary computation to machine learning and adaptive behavior problems. Sine then, the LCS paradigm has broadened greatly into a framework that encompasses many representations, rule discovery mechanisms, and credit assignment schemes. Current LCS applications range from data mining, to automated innovation and the on-line control of cognitive systems. LCS research includes various actual system approaches: While Wilson’s accuracy-based XCS system (1995) has received the highest attention and gained the highest reputation, studies and developments of other LCSs are usually discussed and contrasted.
Advances in machine learning, and reinforcement learning in particular, as well as in evolutionary computation have brought LCS systems the necessary competence and guaranteed learning properties. Novel insights in machine learning and evolutionary computation are being integrated into the LCS framework.
Thus, we invite submissions that discuss recent developments in all areas of research on, and applications of, Learning Classifier Systems. IWLCS is the event that brings together most of the core researchers in classifier systems. Moreover, a free introductory tutorial on LCSs is presented the day before the workshop at GECCO 2008. Tutorial and IWLCS workshop thus also provide an opportunity for researchers interested in LCSs to get an impression of the current research directions in the field as well as a guideline for the application of LCSs to their problem domain.

Submissions and Publication

Submissions will be short-papers up to 8 pages in ACM format. Please see the GECCO 2008 information for authors for further details.
All accepted papers will be presented at IWLCS 2008 and will appear in the GECCO workshop volume. Proceedings of the workshop will be published on CD-ROM, and distributed at the conference. Authors will be invited after the workshop to submit revised (full) papers for publication in the next post-workshop proceedings volume (scheduled for 2009), in the Springer LNCS/LNAI book series.

All papers should be submitted in PDF format and e-mailed to: esterb[at]salle.url.edu.

Important dates

  • Paper submission deadline: April 4, 2008
  • Notification to authors: April 11, 2008
  • Submission of camera-ready material: by Friday, April 18, 2008
  • Conference registration by Monday, April 21, 2008
  • Workshop date: Sunday, July 13, 2008

Committees
Organizing Committee

  • Jaume Bacardit, University of Nottingham (UK). E-mail: jaume.bacardit[at]nottingham.ac.uk
  • Ester Bernadó-Mansilla, Universitat Ramon Llull (Spain). E-mail: esterb[at]salle.url.edu
  • Martin V. Butz, Universitat Wurzburg (Germany). E-mail: mbutz[at]psychologie.uni-wuerzburg.de
Advisory Committee

Alwyn Barry changing jobs

This is an excerpt from Alwyn Barry‘s web page (Thanks Pier Luca for pointing it out)

Most people who know me will be aware that I am changing job shortly. I will be leaving the University of Bath from 30th September 2007, and will no longer be an academic. I am moving to Street in Somerset to become the Pastor of Street Baptist Church. I am still happy to answer any questions relating to my previous research, so do feel free to contact me via my new email address, which is linked from this site.

Alwyn, it has been a pleasure to be able to interact with you. I would like to wish you the best in your new endeavor, knowing that you will give your 100% as usual.

International Workshop on Learning Classifier Systems (IWLCS 2007)

The Tenth International Workshop on Learning Classifier Systems (IWLCS 2007)
will be held on July 7th or 8th, 2007 in association with the conference The Genetic and Evolutionary Computation Conference: GECCO 2007 held at the University College London, in London, England.

Post-workshop proceedings will be published in Springer’s Lecture Notes in Computer Science / Artificial Intelligence series (LNCS/LNAI).

The call For Papers is available here.

Submission deadline is March, 16th.

Call For Papers: The Tenth International Workshop on Learning Classifier Systems (IWLCS 2007)

Call for Papers for IWLCS 2007

The Tenth International Workshop on Learning Classifier Systems (IWLCS 2007) will be held in London, UK, July 7-8, 2007 during the Genetic and Evolutionary Computation Conference (GECCO-2007), July 7-11, 2007.

Since Learning Classifier Systems (LCSs) were introduced by John H. 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, and to the on-line control of cognitive systems. LCS is a very active area of research that encompasses various system approaches. Wilson’s accuracy-based XCS system has received the highest attention and gained the highest reputation.

LCSs are benefiting from recent advances in machine learning, and reinforcement learning in particular, as well as in evolutionary computation. Novel insights in these two areas are continuously integrated into the LCS framework.

We invite submissions which discuss recent developments in all areas of research on, and applications of, Learning Classifier Systems. IWLCS is the event that brings together most of the core researchers in classifier systems. Moreover, a free introductory tutorial on LCSs is presented at GECCO 2007. The IWLCS workshop gives the opportunity also to researchers interested in LCS to get an impression of the current research directions in the field.

Submissions and Publication

There are two ways to submit papers (deadline March 16, 2007):

  1. short papers (up to 4 pages in ACM format) or
  2. full papers (up to 20 pages in Springer format)

All accepted papers may be presented orally at IWLCS. Accepted short papers will appear in the GECCO workshop volume. Proceedings of the workshop will be published on CD-ROM, and distributed at the conference. Authors of short papers will be invited after the workshop to submit revised (full) papers for publication in the post-workshop proceedings, in Springer LNCS/LNAI book series.

Accepted full papers will be published in the post-workshop proceedings. Authors of accepted full papers will be asked to provide a shorter 4-pages version for publication in the GECCO 2007 workshop proceedings.

The normal route is for authors to submit short papers and produce full papers after IWLCS for the post-workshop proceedings, incorporating feedback from reviewers and delegates. All submissions will be peer reviewed. Reviews of short papers will be mainly to provide feedback to enable the production of an improved full paper.

All papers should be submitted in PDF format and e-mailed to: esterb@salle.url.edu.

Important dates

  • Paper submission deadline: Friday, March 16, 2007
  • Notification to authors: Friday, March 30, 2007
  • GECCO camera-ready material: by Wednesday, April 11, 2007
  • Conference registration: Wednesday, April 11, 2007
  • Workshop date: 7th or 8th July
  • Extended paper submissions for LNCS/LNAI post-workshop proceedings: early fall 2007
  • Notification of acceptance: late fall 2007
  • LNCS/LNAI camera ready material: winter 2007/08

Committees

Organizing Commitee

Advisory Committee

For more information please check here.

Preliminary IWLCS 2007 CFP

London, UK, July 7-9, 2007. To be held during the Genetic and Evolutionary Computation Conference (GECCO-2007), July 7-11, 2007.

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 2007.

The final call for papers can be found here.

Advances at the frontier of LCS: LNCS 4399

“Advances at the frontier of Learning Classifier Systems” has been shipped to Springer for the final stages of editing and printing. The volume is going to be printed as Springer’s LNCS 4399 volume. When we started editing this volume, we faced the choice of organizing the contents in a purely chronological fashion or as a sequence of related topics that help walk the reader across the different areas. In the end we decided to organize the contents by area, breaking a little the time-line. This was not a simple endeavor as we could organize the material using multiple criteria. The taxonomy below is our humble effort to provide a coherent grouping. Needless to say, some works may fall in more than one category. Below, you may find the tentative table of contents of the volume. It may change a little bit, but we will keep you posted as soon as we learn from Springer.

Part I. Knowledge representation

  • 1. Analyzing Parameter Sensitivity and Classifier Representations for Real-valued XCS
    by Atsushi Wada, Keiki Takadama, Katsunori Shimohara, and Osamu Katai
    4399 – 001
  • 2. Use of Learning Classifier System for Inferring Natural Language Grammar
    by Olgierd Unold and Grzegorz Dabrowski
    4399 – 018
  • 3. Backpropagation in Accuracy-based Neural Learning Classifier Systems
    by Toby O’Hara and Larry Bull
    4399 – 026
  • 4. Binary Rule Encoding Schemes: A Study Using The Compact Classifier System
    by Xavier Llorà, Kumara Sastry , and David E. Goldberg
    4399 – 041

Part II. Mechanisms

  • 5. Bloat control and generalization pressure using the minimum description length principle for a Pittsburgh approach Learning Classifier System
    by Jaume Bacardit and Josep Maria Garrell
    4399 – 061
  • 6. Post-processing Clustering to Decrease Variability in XCS Induced Rulesets
    by Flavio Baronti, Alessandro Passaro, and Antonina Starita
    4399 – 081
  • 7. LCSE: Learning Classifier System Ensemble for Incremental Medical Instances
    by Yang Gao, Joshua Zhexue Huang, Hongqiang Rong, and Da-qian Gu
    4399 – 094
  • 8. Effect of Pure Error-Based Fitness in XCS
    by Martin V. Butz , David E. Goldberg, and Pier Luca Lanzi
    4399 – 105
  • 9. A Fuzzy System to Control Exploration Rate in XCS
    by Ali Hamzeh and Adel Rahmani
    4399 – 116
  • 10. Counter Example for Q-bucket-brigade under Prediction Problema
    by Atsushi Wada, Keiki Takadama, and Katsunori Shimohara
    4399 – 130
  • 11. An Experimental Comparison between ATNoSFERES and ACS
    by Samuel Landau, Olivier Sigaud, Sébastien Picault, and Pierre Gérard
    4399 – 146
  • 12. The Class Imbalance Problem in UCS Classifier System: A Preliminary Study
    by Albert Orriols-Puig and Ester Bernadó-Mansilla
    4399 – 164
  • 13. Three Methods for Covering Missing Input Data in XCS
    by John H. Holmes, Jennifer A. Sager, and Warren B. Bilker
    4399 – 184

Part III. New Directions

  • 14. A Hyper-Heuristic Framework with XCS: Learning to Create Novel Problem-Solving Algorithms Constructed from Simpler Algorithmic Ingredients
    by Javier G. Marín-Blázquez and Sonia Schulenburg
    4399 – 197
  • 15. Adaptive value function approximations in classifier systems
    by Lashon B. Booker
    4399 – 224
  • 16. Three Architectures for Continuous Action
    by Stewart W. Wilson
    4399 – 244
  • 17. A Formal Relationship Between Ant Colony Optimizers and Classifier Systems
    by Lawrence Davis
    4399 – 263
  • 18. Detection of Sentinel Predictor-Class Associations with XCS: A Sensitivity Analysis
    by John H. Holmes
    4399 – 276

Part IV. Application-oriented research and tools

  • 19. Data Mining in Learning Classifier Systems: Comparing XCS with GAssist
    by Jaume Bacardit and Martin V. Butz
    4399 – 290
  • 20. Improving the Performance of a Pittsburgh Learning Classifier System Using a Default Rule
    by Jaume Bacardit, David E. Goldberg, and Martin V. Butz
    4399 – 299
  • 21. Using XCS to Describe Continuous-Valued Problem Spaces
    by David Wyatt, Larry Bull, and Ian Parmee
    4399 – 318
  • 22. The EpiXCS Workbench: A Tool for Experimentation and Visualization
    by John H. Holmes and Jennifer A. Sager
    4399 – 343