Timezone: »

 
Poster
Model-Based Opponent Modeling
XiaoPeng Yu · Jiechuan Jiang · Wanpeng Zhang · Haobin Jiang · Zongqing Lu

@

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, naive learner, and reasoning learner.

Keywords: multi-agent reinforcement learning · opponent modeling
[ Paper]  [ Poster]  [ OpenReview] 

Author Information

XiaoPeng Yu (Peking University, Tsinghua University)
Jiechuan Jiang (Peking University)
Wanpeng Zhang (Peking University)
Haobin Jiang (Peking University)
Zongqing Lu (Peking University)

More from the Same Authors