L2E: Learning to Exploit Your Opponent
This work addresses the challenge of exploiting sub-optimal opponents in games like poker and soccer, offering a method that reduces data needs and improves adaptability, though it is incremental in the field of opponent modeling.
The paper tackles the problem of opponent modeling in strategic interactions by proposing the Learning to Exploit (L2E) framework, which uses implicit modeling and a strategy generation algorithm to quickly adapt to unknown opponents with diverse styles, as demonstrated in poker and grid soccer games.
Opponent modeling is essential to exploit sub-optimal opponents in strategic interactions. Most previous works focus on building explicit models to directly predict the opponents' styles or strategies, which require a large amount of data to train the model and lack adaptability to unknown opponents. In this work, we propose a novel Learning to Exploit (L2E) framework for implicit opponent modeling. L2E acquires the ability to exploit opponents by a few interactions with different opponents during training, thus can adapt to new opponents with unknown styles during testing quickly. We propose a novel opponent strategy generation algorithm that produces effective opponents for training automatically. We evaluate L2E on two poker games and one grid soccer game, which are the commonly used benchmarks for opponent modeling. Comprehensive experimental results indicate that L2E quickly adapts to diverse styles of unknown opponents.