Diversity-based Deep Reinforcement Learning Towards Multidimensional Difficulty for Fighting Game AI
In this EXAG 2022 presentation, Emily Halina discusses how in fighting games, individual players of the same skill level often exhibit distinct strategies from one another through their gameplay. Despite this, the majority of AI agents for fighting games have only a single strategy for each ``level'' of difficulty. To make AI opponents more human-like, they'd ideally like to see multiple different strategies at each level of difficulty, a concept they refer to as ``multidimensional'' difficulty. They introduce a diversity-based deep reinforcement learning approach for generating a set of agents of similar difficulty that utilize diverse strategies. They find this approach outperforms a human-authored baseline in both diversity and performance.