Schedule

Saturday 26th June 2004

The talks are 25 min. plus 10 min. for discussion.

Session I (8:30 -10:20)

8:30 - 8:35
Introduction
8:35 - 9:10
Atsushi Wada, Keiki Takadama, Katsunori Shimohara and Osamu Katai:
Is Gradient Descent Method Effective in XCS? Analysis of Reinforcement Process in XCSG
9:10 - 9:45
David LeRoux and Micheal Littman:
Reinforcement Learning using LCS in Continuous State Space
9:45 - 10:20
Stewart W. Wilson:
Optimal Continous Policies: A Classifier System Approach

Session II (10:40 - 12:30)

10:40 - 11:15
Martin V. Butz and David E. Goldberg:
Hierarchical Classification Problems Demand Effective Building Block Identification and Processing in LCSs
11:15 -11:50
Olivier Sigaud and Samuel Landau:
A Michigan style architecture for learning finite state controllers: a first step
11:50 - 12:25
Tim Kovacs and Chuan Yang:
An Initial Study of a Model-Based XCS

Session III (14:00 - 15:50)

14:00 - 14:35
John H. Holmes, Jennifer A. Sager and Warren B. Bilker:
A Comparison of Three Methods for Covering Missing Data in XCS
14:35 - 15:10
Jaume Bacardit and Martin V. Butz:
Data Mining in Learning Classifier Systems: Comparing XCS with GAssist
15:10 - 15:45
Jaume Bacardit, David E. Goldberg and Martin V. Butz:
Improving the Performance of a Pittsburgh Learning Classifier System Using a Default Rule

Session IV (16:10 - 18:00)

General discussion about the future of IWLCS