CHUNKING FOR EXPERIENCE Michael George Jonathan Schaeffer Department of Computing Science University of Alberta Edmonton, Alberta CANADA T6G 2H1 Abstract Human game players rely heavily on the experience gained by playing over the games of masters. A player may recall a previous game to either obtain the best move (if he has previously seen the identical position) or suggest a best move (if similar to others seen). However, game-playing programs operate in isolation, relying on the combination of search and programmed knowledge to discover the best move, even in positions well-known to humans. At best, programs have only a limited amount of information about previous games. This paper discusses enhancing a chess-playing program to discover and extract implicit knowledge from previously played grandmaster games, and using it to improve the chess program’s performance. During a game, a database of positions is queried looking for identical or similar posi- tions to those on the board. Similarity measures are determined by chunking the position and using these patterns as indices into the database. Relevant information is subsequently passed back to the chess program and used in its decision making process. As the number of games in the database increases, the "experience" available to the program improves the likelihood that relevant, useful information can be found for a given position. 1. Introduction For many subject areas requiring skills, a large body of previous experience is usu- ally gathered which can be used as a resource to be studied for improving one’s perfor- mance. Unfortunately, most attempts at having machines mimic human behavior are pri- marily concerned with trying to solve the problem in isolation (usually a difficult enough task), without exploiting the available wealth of others’ experience. For many discip- lines, this information is not available in machine readable form. Other than using simple memory tables, few programs attempt to take advantage of the acquired wealth of human experience on a particular subject. For most games, a player’s strength can be increased by re-playing and analyzing hundreds of master-level games. Besides the obvious skills that this can improve, a signi- ficant side-effect is the human’s ability to remember the games or important fragments from them. During a game, strong players are frequently able to relate the position on the board to an identical or similar one seen in a game played by themselves or by others.