Essay.10-ZhancunMu-2100017790
Keywords: Explainable AI, Explainable reinforcement learning, Causality
Abstract: Explainable reinforcement learning (XRL) aims to provide transparency into the behavior and decision-making of RL agents. As RL systems grow in capability and autonomy, model interpretability becomes crucial. This essay discusses approaches in XRL from the perspective of causality. Learning causality provides opportunities to open the black box of RL agents.
Submission Number: 212
Loading