Performance of Monte Carlo Tree Search Algorithms when Playing the Game Ataxx

Abstract

Monte Carlo Tree Search (MCTS) has recently emerged as a promising technique to play games with very large state spaces. Ataxx is a simple two-player board game with large and deep game tree. In this work, we apply different MCTS algorithms to play the game Ataxx and evaluate its performance against different adversaries (e.g., minimax2). Our analysis highlights one key aspect of MCTS, the trade-off between samples (and accuracy) and chances of winning the game which translates to a trade-off between the delay in making a move and chances of winning.

Related Papers

• → Monte-Carlo tree search with Epsilon-Greedy for game of amazons(2023)1 cited
• → High-Level Representations for Game-Tree Search in RTS Games(2014)15 cited
• → Deep learning for general game playing with Ludii and Polygames(2022)12 cited
• → Solving Connect 4 Using Artificial Intelligence(2021)2 cited
• → Deep Learning for General Game Playing with Ludii and Polygames(2021)