🤖 AI Summary
This work addresses the limited interpretability of deep reinforcement learning policies in locomotion control, despite their strong performance. Focusing on policies trained in the HalfCheetah-v5 environment, the study identifies semantically meaningful phases through clustering based on state similarity and transition consistency, revealing that neural networks can spontaneously develop human-interpretable periodic phase structures and branching logic without explicit supervision. To further elucidate these emergent behaviors, the authors employ Explainable Boosting Machines (EBM) to model each phase, explicitly uncovering the critical state features and action rules governing phase-specific control. This approach substantially enhances policy interpretability and offers a novel perspective for understanding the implicit motion planning strategies learned by autonomous agents.
📝 Abstract
In locomotion control tasks, Deep Reinforcement Learning (DRL) has demonstrated high performance; however, the decision-making process of the learned policy remains a black box, making it difficult for humans to understand. On the other hand, in periodic motions such as walking, it is well known that implicit motion phases exist, such as the stance phase and the swing phase. Focusing on this point, this study hypothesizes that a policy trained for locomotion control may also represent a phase structure that is interpretable by humans. To examine this hypothesis in a controlled setting, we consider a locomotion task that is amenable to observing whether a policy autonomously acquires temporally structured phases through interaction with the environment. To verify this hypothesis, in the MuJoCo locomotion benchmark HalfCheetah-v5, the state transition sequences acquired by a policy trained for walking control through interaction with the environment were aggregated into semantic phases based on state similarity and consistency of subsequent transitions. As a result, we demonstrated that the state sequences generated by the trained policy exhibit periodic phase transition structures as well as phase branching. Furthermore, by approximating the states and actions corresponding to each semantic phase using Explainable Boosting Machines (EBMs), we analyzed phase-dependent decision making-namely, which state features the policy function attends to and how it controls action outputs in each phase. These results suggest that neural network-based policies, which are often regarded as black boxes, can autonomously acquire interpretable phase structures and logical branching mechanisms.