ORCAID: Oblique Rule-Based Continuous-Action Interpretation for Deep RL Policies
For RL practitioners needing interpretable policies in continuous-control tasks, ORCAID offers a novel method that balances performance and interpretability.
ORCAID extracts interpretable rule-based policies from deep RL agents in continuous action spaces using oblique decision trees, achieving strong performance with low parameter counts and even improving the original policy.
Explainability remains a key issue in reinforcement learning (RL). Distilling an interpretable policy from an agent trained in a complex environment is particularly challenging when the action space is continuous. We introduce ORCAID, a novel method for extracting interpretable rule-based policies from RL agents operating in mixed continuous-discrete environments with continuous action spaces. Our main contribution is an efficient oblique decision tree training algorithm that partitions the state space by hyperplanes and fits local linear models. The key idea lies in a three-stage split search: efficient random initialization, local refinement, and backward elimination. Finally, adjacent leaves are merged to yield a concise set of interpretable rules describing a given deep RL policy. We evaluate ORCAID across multiple RL environments, demonstrating that the extracted rule-based policies maintain strong performance with a low number of parameters and can even be used to improve the performance of the original deep RL policy.