Optimization of Connect6 based on Principal Variation Search and Transposition Tables Algorithms

Abstract

Connect6 is an intuitive and fair game with no first-hand advantage. Because of the large search tree, the search depth is often lower, to sacrifice strength of chess to improve search efficiency. In this case, this paper proposes an optimization algorithm for pruning Connect6 search tree based on the principal variation search and transposition tables algorithms. The optimization algorithm obtained PV node through searching, narrowed the pruning interval, and stored the key value of the situation into the transposition table through searching, and then directly returned the evaluation value when the same chess game was searched again, which greatly improved the search efficiency. Through a large number of repeated experiments, the effectiveness of the algorithm can be proved.