Domain Generalization for Robust Model-Based Offline Reinforcement Learning
This addresses a practical problem in offline RL for scenarios with multiple human demonstrators, offering an incremental improvement over existing methods.
The paper tackles the problem of multi-demonstrator offline reinforcement learning, where data from multiple human operators is used without assumptions about their policies, by framing it as a domain generalization issue and applying Risk Extrapolation to learn dynamics and rewards models, resulting in improved domain generalization and superior policies in offline model-based RL.
Existing offline reinforcement learning (RL) algorithms typically assume that training data is either: 1) generated by a known policy, or 2) of entirely unknown origin. We consider multi-demonstrator offline RL, a middle ground where we know which demonstrators generated each dataset, but make no assumptions about the underlying policies of the demonstrators. This is the most natural setting when collecting data from multiple human operators, yet remains unexplored. Since different demonstrators induce different data distributions, we show that this can be naturally framed as a domain generalization problem, with each demonstrator corresponding to a different domain. Specifically, we propose Domain-Invariant Model-based Offline RL (DIMORL), where we apply Risk Extrapolation (REx) (Krueger et al., 2020) to the process of learning dynamics and rewards models. Our results show that models trained with REx exhibit improved domain generalization performance when compared with the natural baseline of pooling all demonstrators' data. We observe that the resulting models frequently enable the learning of superior policies in the offline model-based RL setting, can improve the stability of the policy learning process, and potentially enable increased exploration.