phil@nte-scg.UUCP (Phil Trubey) (08/21/85)
I just ran across a 1978 article on a program that plays Go. It's called... "Pattern Recognition and Pattern-Directed Inference In A Program For Playing Go" by Walter Reitman and Bruce Wilcox of the University of Michigan. Found in _Pattern_Directed_Inference_Systems_ edited by D.A. Waterman and F. Hayes-Roth. Academic Press, 1978. It's an interesting article that describes the general method they used. The program that they had come up with played with a strength of about 27 kyu ... about equivalent to a person who had played about 40-50 previous games in his life. Anyways ... does anybody know if this research was followed up in any way? Have there been other (recent) attempts at constructing computer Go players? ... on micros maybe? The program refered to above ran on an Amdahl/470. -- Phil Trubey Northern Telecom Electronics Ltd. Ottawa, Ontario Mail path: ...!bnr-vpa!nte-scg!phil