Previous Issues

ABSTRACT

Researchers in game playing have developed a class of practical algorithms called alpha-beta pruning to improve the efficiency of search in two-player games. However, these efficient algorithms appear to be inadequate to capture all necessary information to play the game of Go at a level above that of an intermediate human player. This is due to the complexity and knowledge-based requirement of this game. In this paper, search and knowledge are both involved in solving one of the resource-demanding non-trivial artificial intelligence applications. Together, they constitute a basic skeleton of a program that is capable of playing a complete game of Go. This program is designed to reflect on the significance of adding knowledge heuristics in the form of pattern matching and book moves to the search technique offered by the traditional alpha-beta algorithm.