Does it matter how well I know what you're thinking? Opponent Modelling in an RTS game

TL;DR

This paper examines how the accuracy of opponent modeling affects the performance of MCTS and RHEA in an RTS game, finding MCTS more robust to inaccuracies and recommending modeling strategies accordingly.

Contribution

It compares the sensitivity of MCTS and RHEA to opponent model accuracy in an RTS game, highlighting MCTS's robustness and providing practical modeling recommendations.

Findings

MCTS performs better with inaccurate models than RHEA.

RHEA's performance degrades significantly with inaccurate models.

Modeling opponent actions within MCTS is preferable under low computational budgets.

Abstract

Opponent Modelling tries to predict the future actions of opponents, and is required to perform well in multi-player games. There is a deep literature on learning an opponent model, but much less on how accurate such models must be to be useful. We investigate the sensitivity of Monte Carlo Tree Search (MCTS) and a Rolling Horizon Evolutionary Algorithm (RHEA) to the accuracy of their modelling of the opponent in a simple Real-Time Strategy game. We find that in this domain RHEA is much more sensitive to the accuracy of an opponent model than MCTS. MCTS generally does better even with an inaccurate model, while this will degrade RHEA's performance. We show that faced with an unknown opponent and a low computational budget it is better not to use any explicit model with RHEA, and to model the opponent's actions within the tree as part of the MCTS algorithm.