Some Studies in Mucki ~ Learning Using the Game of Checkers

-A new signature-table technique is described together with an improved book-learning procedure which is thought to be much superior to the linear polynomial method. Full use is made of the so-cahed “alpha-beta” pruning and several forms of forward pruning to restrict the spread of the move tree and to permit the program to look ahead to a much greater depth than it otherwise cotdd do. While still unable to outplay checker masters, the program’s playing ability has been greatly improved. Limited progress has been made in the development of an improved book-learning technique and in the optimization of playing strategies as applied to the checker playing program described in an earlier paper with this same title.7 Because of the sharpening in our understanding and the substantial improvements in playing ability that have resulted from these recent studies, a reporting at this time seems desirable. Unfortunately, the most basic limitation of the known machine learning techniques, as previously outlined, has not yet been overcome, nor has the program been able to outplay the best human checker players.1 We will briefly review the earlier work. The reader who does not find this review adequate might do well to refresh his memory by referring to the earlier paper. Two machine learning procedures were described in some detail: (1) a rote learning procedure in which a record was kept of the board t Some Studies in Machine Learning Using the Game of Checkam, IBM Journal 3, 211-299 (1959). Reprinted (with minor additions and cormctions) in COPQJ~&~S and ?Yro#gJrr, edited by Feigenbaum and Feldman, McGraw-Hi 1963. $ In a 1965 match with the program, the World Champion, Mr. W. F. Hellmao, won all four games played by mail but was played to a draw in one hurriedly played cross-board game. Rezently Mr. K. D. Hanson, the F’acific. Coast Champion, has beaten current versions of the program on two separate occasions.