Boolean Decision Rules for Reinforcement Learning Policy Summarisation
This work addresses the challenge of making RL policies more interpretable for safety-critical applications, though it is incremental as it builds on existing explainability methods.
The paper tackles the problem of explainability in Reinforcement Learning policies by proposing a Boolean Decision Rules model to create post-hoc rule-based summaries, demonstrating its application on a DQN agent in a lava gridworld to generate simple generalized rules.
Explainability of Reinforcement Learning (RL) policies remains a challenging research problem, particularly when considering RL in a safety context. Understanding the decisions and intentions of an RL policy offer avenues to incorporate safety into the policy by limiting undesirable actions. We propose the use of a Boolean Decision Rules model to create a post-hoc rule-based summary of an agent's policy. We evaluate our proposed approach using a DQN agent trained on an implementation of a lava gridworld and show that, given a hand-crafted feature representation of this gridworld, simple generalised rules can be created, giving a post-hoc explainable summary of the agent's policy. We discuss possible avenues to introduce safety into a RL agent's policy by using rules generated by this rule-based model as constraints imposed on the agent's policy, as well as discuss how creating simple rule summaries of an agent's policy may help in the debugging process of RL agents.