Search papers, labs, and topics across Lattice.
This paper investigates whether locomotion policies learned via DRL autonomously acquire interpretable phase structures, similar to the stance and swing phases in walking. They trained a policy on HalfCheetah-v5 and then aggregated state transition sequences into semantic phases based on state similarity and transition consistency. The results show that the learned policy exhibits periodic phase transitions and branching logic, and these phases can be approximated with Explainable Boosting Machines to understand phase-dependent decision making.
Locomotion policies, often considered black boxes, can autonomously learn interpretable phase structures and branching logic, revealing a hidden order in their decision-making.
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.
