Learning Through the KRKa2 Chess Ending
The chess ending (KRKa2) has been studied using decision trees, neural networks and human reasoning to build a classifier for this ending, and for the discovery of convenient chess attributes. These chess attributes will serve for testing new ideas in planning. The idea is to investigate whether good automatically learnt policies for a planning problem can be generated using training examples along with evolutionary algorithms. The training examples, used as input to the learning algorithm, describe specific descriptions of a number of solved instances in one domain; then to improve the learnt policies obtained from the training examples, the policies should evolve. We believe that the domain of games is well-suited for testing these new ideas in planning.
Unable to display preview. Download preview PDF.
- 1.Bain, M.E.: Learning Logical Exceptions in Chess. Ph. D. thesis, Department of Statistics and Modelling Science, University of Strathclyde, Scotland (1994)Google Scholar
- 2.De Groot, A., Gobet, F.: Perception and memory in chess. In: Jongman, R.W. (ed.) Heuristics of the professional eye, Assen, Van Gorcum (1996)Google Scholar
- 3.Fürnkranz, J.: Machine learning in computer chess: The next generation. International Computer Chess Association Journal 19(3) (1996)Google Scholar
- 4.Khardon, R.: Learning action strategies for planning domains. Artificial Intelligence 113 (1999)Google Scholar
- 5.González, A.: Learning Policy Sets using Evolutionary Computation (2003), http://www.lsi.upc.es/~alquezar/gonzalez2003.ps
- 6.Shapiro, A.D., Niblett, T.: Automatic induction of classification rules for a chess endgame. In: Clarke, M.R.B. (ed.) Advances in Computer Chess, vol. 3, pp. 73–92. Pergamon Press, Oxford (1983)Google Scholar
- 7.Shapiro, A.D.: Structured Induction in Expert Systems. Turing Institute Press. Addison-Wesley (1987)Google Scholar