A Survey of Monte Carlo Tree Search Methods

TLDR

Monte Carlo tree search (MCTS) blends tree‑search precision with random sampling generality, has achieved spectacular success in computer Go and other domains, and faces open research questions that signal future work. This survey aims to provide a snapshot of the state of the art in MCTS after its first five years of research. The authors outline the core algorithm’s derivation, categorize its many variations and enhancements, and describe how these are applied across domains. They summarize performance results from key game and non‑game domains, illustrating MCTS’s effectiveness and the impact of its variations.

Abstract

Monte Carlo tree search (MCTS) is a recently proposed search method that combines the precision of tree search with the generality of random sampling. It has received considerable interest due to its spectacular success in the difficult problem of computer Go, but has also proved beneficial in a range of other domains. This paper is a survey of the literature to date, intended to provide a snapshot of the state of the art after the first five years of MCTS research. We outline the core algorithm's derivation, impart some structure on the many variations and enhancements that have been proposed, and summarize the results from the key game and nongame domains to which MCTS methods have been applied. A number of open research questions indicate that the field is ripe for future work.