Optimizing GoTools' Search Heuristics using Genetic Algorithms

TL;DR

This paper presents a method to enhance GoTools' search heuristics by applying genetic algorithms, leading to improved problem-solving efficiency in the game of Go.

Contribution

It introduces a genetic algorithm-based approach to optimize heuristic weights in GoTools and implements a parallel MPI interface for efficient fitness evaluations.

Findings

Optimized heuristic weights improve GoTools' performance.

Parallel MPI implementation accelerates the optimization process.

Framework supports future extension for more heuristic parameters.

Abstract

GoTools is a program which solves life & death problems in the game of Go. This paper describes experiments using a Genetic Algorithm to optimize heuristic weights used by GoTools' tree-search. The complete set of heuristic weights is composed of different subgroups, each of which can be optimized with a suitable fitness function. As a useful side product, an MPI interface for FreePascal was implemented to allow the use of a parallelized fitness function running on a Beowulf cluster. The aim of this exercise is to optimize the current version of GoTools, and to make tools available in preparation of an extension of GoTools for solving open boundary life & death problems, which will introduce more heuristic parameters to be fine tuned.