Explainable Reinforcement Learning via Temporal Policy Decomposition
This addresses the problem of interpretability in RL for researchers and practitioners, offering a method to enhance transparency in sequential decision-making, though it is incremental as it builds on existing value function frameworks.
The paper tackles the challenge of explaining Reinforcement Learning policies by introducing Temporal Policy Decomposition (TPD), which decomposes value functions into Expected Future Outcomes to clarify actions based on temporal sequences, resulting in explanations that improve understanding of policy strategies and reward composition.
We investigate the explainability of Reinforcement Learning (RL) policies from a temporal perspective, focusing on the sequence of future outcomes associated with individual actions. In RL, value functions compress information about rewards collected across multiple trajectories and over an infinite horizon, allowing a compact form of knowledge representation. However, this compression obscures the temporal details inherent in sequential decision-making, presenting a key challenge for interpretability. We present Temporal Policy Decomposition (TPD), a novel explainability approach that explains individual RL actions in terms of their Expected Future Outcome (EFO). These explanations decompose generalized value functions into a sequence of EFOs, one for each time step up to a prediction horizon of interest, revealing insights into when specific outcomes are expected to occur. We leverage fixed-horizon temporal difference learning to devise an off-policy method for learning EFOs for both optimal and suboptimal actions, enabling contrastive explanations consisting of EFOs for different state-action pairs. Our experiments demonstrate that TPD generates accurate explanations that (i) clarify the policy's future strategy and anticipated trajectory for a given action and (ii) improve understanding of the reward composition, facilitating fine-tuning of the reward function to align with human expectations.