Optimizing Selective Search in Chess

TL;DR

This paper presents a genetic algorithm-based method for automatically tuning search parameters in chess programs, achieving performance comparable to top manually tuned systems.

Contribution

Introduces a novel genetic algorithm approach for automatic parameter tuning in chess engines, reducing reliance on manual tuning.

Findings

Automated parameter tuning matches top tournament-level performance.

Genetic algorithms effectively optimize complex search parameters.

Automatic tuning simplifies development of high-performing chess programs.

Abstract

In this paper we introduce a novel method for automatically tuning the search parameters of a chess program using genetic algorithms. Our results show that a large set of parameter values can be learned automatically, such that the resulting performance is comparable with that of manually tuned parameters of top tournament-playing chess programs.