General Game Heuristic Prediction Based on Ludeme Descriptions

TL;DR

This paper evaluates heuristics for general game playing in Ludii, trains models to predict heuristic performance from game descriptions, and analyzes the Ludii game set and its defining ludemes.

Contribution

It introduces a method to predict heuristic effectiveness from game descriptions and provides an analysis of Ludii's game content and ludemes.

Findings

Regression models can predict heuristic performance from game descriptions.

Analysis reveals key ludemes that define Ludii games.

Performance varies across different heuristics and game types.

Abstract

This paper investigates the performance of different general-game-playing heuristics for games in the Ludii general game system. Based on these results, we train several regression learning models to predict the performance of these heuristics based on each game's description file. We also provide a condensed analysis of the games available in Ludii, and the different ludemes that define them.