Towards a Characterisation of Monte-Carlo Tree Search Performance in Different Games

TL;DR

This paper presents a large dataset of Monte-Carlo Tree Search (MCTS) performance across numerous games, aiming to understand which variants work best in different contexts and to develop predictive models.

Contribution

It introduces an extensive dataset of MCTS agent performances in various games and explores initial analysis and predictive modeling efforts.

Findings

Initial analysis of MCTS variants across games

Development of predictive models for MCTS performance

Lessons learned for future dataset improvements

Abstract

Many enhancements to Monte-Carlo Tree Search (MCTS) have been proposed over almost two decades of general game playing and other artificial intelligence research. However, our ability to characterise and understand which variants work well or poorly in which games is still lacking. This paper describes work on an initial dataset that we have built to make progress towards such an understanding: 268,386 plays among 61 different agents across 1494 distinct games. We describe a preliminary analysis and work on training predictive models on this dataset, as well as lessons learned and future plans for a new and improved version of the dataset.