Research map / Reinforcement learning / Visual and representation RL

UTILITY: Utilizing Explainable Reinforcement Learning to Improve Reinforcement Learning

ICLR 2025 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Uses explainable RL (XRL) techniques to diagnose and improve the performance of a trained RL agent, aiming to address non-optimal agents.

Problem

Trained RL agents are often non-optimal and far from optimal, and they lack explainability.

Approach

Repurposes explainable RL, which was built to explain agent decisions, as a means to improve RL performance itself rather than only to interpret the agent.

Closest papers by meaning

• Explaining Reinforcement Learning Agents through Counterfactual Action Outcomes (AAAI 2024 · similarity 0.89)
• Explanations for Multi-Agent Reinforcement Learning (AAAI 2025 · similarity 0.89)
• ORCAID: Oblique Rule-Based Continuous-Action Interpretation for Deep RL Policies (NeurIPS 2026 · similarity 0.86)
• RICE: Breaking Through the Training Bottlenecks of Reinforcement Learning with Explanation (ICML 2024 · similarity 0.86)
• The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models (NeurIPS 2026 · similarity 0.86)
• Learning from Ambiguous Demonstrations with Self-Explanation Guided Reinforcement Learning (AAAI 2024 · similarity 0.86)