THE EVOLUTION OF STRONG OTHELLO PROGRAMS

摘要

This paper surveys the evaluation and search techniques utilized by the strongest Othello programs of their time during the past twenty years. In this time span computer Othello has experienced considerable progress which culminated in the convincing 6-0 victory of LOGISTELLO against the then World-champion Takeshi Murakami in 1997. The focus of this article is the evolution of Othello evaluation functions and heuristic search techniques which quite nicely reflect the general A.I. trend of replacing slow and error-prone manual tuning by automated machine learning approaches.