Boosting Offline Reinforcement Learning with Residual Generative Modeling
This work addresses a neglected issue in offline RL for improving policy learning in domains such as gaming, though it is incremental as it builds on existing generative modeling approaches.
The paper tackles the problem of error propagation in generative modeling for offline reinforcement learning, proposing a residual generative model called AQL that learns more accurate policy approximations, achieving competitive performance in complex control tasks like the MOBA game Honor of Kings.
Offline reinforcement learning (RL) tries to learn the near-optimal policy with recorded offline experience without online exploration. Current offline RL research includes: 1) generative modeling, i.e., approximating a policy using fixed data; and 2) learning the state-action value function. While most research focuses on the state-action function part through reducing the bootstrapping error in value function approximation induced by the distribution shift of training data, the effects of error propagation in generative modeling have been neglected. In this paper, we analyze the error in generative modeling. We propose AQL (action-conditioned Q-learning), a residual generative model to reduce policy approximation error for offline RL. We show that our method can learn more accurate policy approximations in different benchmark datasets. In addition, we show that the proposed offline RL method can learn more competitive AI agents in complex control tasks under the multiplayer online battle arena (MOBA) game Honor of Kings.