Model-Based Opponent Modeling
This work addresses the problem of opponent adaptation in multi-agent systems for AI agents, representing an incremental improvement by integrating multiple opponent types into a single framework.
The paper tackles the challenge of adapting to diverse unseen opponents in multi-agent environments by proposing model-based opponent modeling (MBOM), which simulates recursive reasoning and mixes imagined policies, achieving more effective adaptation than existing methods across tasks with fixed, naïve learner, and reasoning learner opponents.
When one agent interacts with a multi-agent environment, it is challenging to deal with various opponents unseen before. Modeling the behaviors, goals, or beliefs of opponents could help the agent adjust its policy to adapt to different opponents. In addition, it is also important to consider opponents who are learning simultaneously or capable of reasoning. However, existing work usually tackles only one of the aforementioned types of opponents. In this paper, we propose model-based opponent modeling (MBOM), which employs the environment model to adapt to all kinds of opponents. MBOM simulates the recursive reasoning process in the environment model and imagines a set of improving opponent policies. To effectively and accurately represent the opponent policy, MBOM further mixes the imagined opponent policies according to the similarity with the real behaviors of opponents. Empirically, we show that MBOM achieves more effective adaptation than existing methods in a variety of tasks, respectively with different types of opponents, i.e., fixed policy, naïve learner, and reasoning learner.