Search papers, labs, and topics across Lattice.
This paper outlines a novel evaluation benchmark, EvalXRL, for assessing Explainable Reinforcement Learning (XRL) methods based on their effectiveness in diagnosing and fixing bugs in RL agents. By employing a Large Language Model (LLM) coding agent, the study iteratively tests various XRL methods across different environments and malfunctions, using the agents' reward signals to quantify performance. The findings highlight the potential of closed-loop evaluation to enhance the practical utility of XRL methods, positioning them as tools for real-world debugging rather than mere theoretical constructs.
Evaluating XRL methods by their ability to fix RL agent bugs reveals significant differences in their practical effectiveness, challenging existing assessment paradigms.
This preliminary paper outlines a planned evaluation benchmark for Explainable Reinforcement Learning (XRL) methods. Current evaluations rely on functionally-grounded metrics like faithfulness and compactness, and on human-grounded proxies like subjective ratings or prediction accuracy. We suggest evaluating XRL methods by how effectively their generated explanations help to diagnose and fix malfunctioning reinforcement learning (RL) agents. We propose EvalXRL, a benchmark in which a Large Language Model (LLM) coding agent uses different XRL methods to diagnose a held-out malfunction in an RL agent, and then repair it. Our proposed benchmark iterates across (environment $\times$ malfunction $\times$ XRL method) tuples and uses the reward signal of the RL agents to form a final score for each XRL method. The coding agent may use the method interactively: invoke the XRL method, process its output, form new hypotheses on what is broken, and invoke the method again with parameters adjusted for testing these hypotheses. This closed-loop structure may be described as a simplified version of the scientific method. Some XRL methods provide self-evaluations that follow this pattern; we propose the first head-to-head comparison of multiple XRL methods in closed-loop usage.
