Effect of Potential Model Pruning on Different-Sized Boards in Monte-Carlo GO

Abstract

Monte-Carlo GO is a computer GO program that is sufficiently competent without the knowledge expressions of IGO. Although it is computationally intensive, the computational complexity can be reduced by properly pruning the IGO game tree. In this study, we achieved this by using a potential model based on the knowledge expressions of IGO. The potential model treats GO stones as potentials. A specific potential distribution on the GO board results from a unique arrangement of stones on the board. Pruning with the potential model categorizes legal moves into effective and ineffective moves in accordance with the potential threshold. In this study, certain pruning strategies based on potentials and potential gradients were experimentally evaluated. In particular, for different-sized boards, the effects of pruning strategies were evaluated in terms of their robustness. We successfully demonstrated pruning with a potential model to reduce the computational complexity of the game of GO as well as the robustness of this effect across different-sized boards.