Monte-Carlo Tree Search for Implementation of Dynamic Difficulty Adjustment Fighting Game AIs Having Believable Behaviors

Abstract

In this paper, we propose a Monte-Carlo Tree Search (MCTS) fighting game AI capable of dynamic difficulty adjustment while maintaining believable behaviors. This work targets beginner-level and intermediate-level players. In order to improve players’ skill while at the same time entertaining them, AIs are needed that can evenly fight against their opponent beginner and intermediate players, and such AIs are called dynamic difficulty adjustment (DDA) AIs. In addition, in order not to impair the players’ playing motivation due to the AI’s unnatural actions such as intentionally taking damage with no resistance, DDA methods considering restraint of its unnatural actions are needed. In this paper, for an MCTS-based AI previously proposed by the authors’ group, we introduce a new evaluation term on action believability, to the AIs evaluation function, that focuses on the amount of damage to the opponent. In addition, we introduce a parameter that dynamically changes its value according to the current game situation in order to balance this new term with the existing one, focusing on adjusting the AI’s skill equal to that of the player, in the evaluation function. Our results from the conducted experiment using FightingICE, a fighting game platform used in a game AI competition at CIG since 2014, show that the proposed DDA-AI can dynamically adjust its strength to its opponent human players, especially intermediate players, while restraining its unnatural actions throughout the game.