ASQ-IT: Interactive Explanations for Reinforcement-Learning Agents
This addresses the need for better explainability in reinforcement learning for end-users, offering an interactive approach rather than static explanations, though it is incremental in combining formal methods with user interaction.
The paper tackles the problem of making reinforcement learning agents more interpretable by introducing ASQ-IT, an interactive tool that allows users to query agent behaviors using temporal logic, resulting in users being able to identify faulty behaviors effectively.
As reinforcement learning methods increasingly amass accomplishments, the need for comprehending their solutions becomes more crucial. Most explainable reinforcement learning (XRL) methods generate a static explanation depicting their developers' intuition of what should be explained and how. In contrast, literature from the social sciences proposes that meaningful explanations are structured as a dialog between the explainer and the explainee, suggesting a more active role for the user and her communication with the agent. In this paper, we present ASQ-IT -- an interactive tool that presents video clips of the agent acting in its environment based on queries given by the user that describe temporal properties of behaviors of interest. Our approach is based on formal methods: queries in ASQ-IT's user interface map to a fragment of Linear Temporal Logic over finite traces (LTLf), which we developed, and our algorithm for query processing is based on automata theory. User studies show that end-users can understand and formulate queries in ASQ-IT, and that using ASQ-IT assists users in identifying faulty agent behaviors.