A model of visual organization for the game of GO

No successful GO-playing program has appeared in the literature, although Remus used GO as the subject of a machine learning study, and Thorp and Walden have considered some of its mathematical aspects. Another author considered GO to be somewhat mysterious, making it a challenge to those interested in automating it. Apparently the game was described as being mysterious to indicate that people were able to play it without knowing how they were able to play so well. More study of this complex game may reward us with new insight into human perceptual and problem solving abilities as well as foster the development of new techniques for artificial intelligence. This report describes a program which plays GO. The program uses an information processing model to produce perceptual features which are seen by human GO players, and is capable of several responses to the recognition of significant configurations of these perceptual features.