Towards a Characterisation of Monte-Carlo Tree Search Performance in Different Games
This work addresses a gap in AI research for game playing by providing a dataset to characterize MCTS performance, but it is incremental as it focuses on initial data collection and analysis without major breakthroughs.
The paper tackled the problem of understanding which Monte-Carlo Tree Search variants perform well in different games by building an initial dataset of 268,386 plays among 61 agents across 1,494 games, and conducted preliminary analysis and model training to progress towards this goal.
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.