A self-learning evolutionary chess program

Author

Fogel, David B. ; Hays, Timothy J. ; Hahn, Sarah L. ; Quon, James

Author_Institution

Natural Selection Inc., La Jolla, CA, USA

Volume

92

Issue

12

fYear

2004

fDate

12/1/2004 12:00:00 AM

Firstpage

1947

Lastpage

1954

Abstract

A central challenge of artificial intelligence is to create machines that can learn from their own experience and perform at the level of human experts. Using an evolutionary algorithm, a computer program has learned to play chess by playing games against itself. The program learned to evaluate chessboard configurations by using the positions of pieces, material and positional values, and neural networks to assess specific sections of the chessboard. During evolution, the program improved its play by almost 400 rating points. Testing under simulated tournament conditions against Pocket Fritz 2.0 indicated that the evolved program performs above the master level.

Keywords

computer games; evolutionary computation; neural nets; unsupervised learning; artificial intelligence; chessboard configuration; computer games; evolutionary algorithm; evolutionary chess program; neural networks; pocket Fritz 2.0; self learning; Artificial intelligence; Computational modeling; Evolutionary computation; Feedback; Hardware; Humans; Machine learning; Machine learning algorithms; Neural networks; Testing; Chess; computational intelligence; evolutionary computation; machine learning; neural networks;

fLanguage

English

Journal_Title

Proceedings of the IEEE

Publisher

ieee

ISSN

0018-9219

Type

jour

DOI

10.1109/JPROC.2004.837633

Filename

1360168