Explainable Reinforcement Learning Through Goal-Based Explanations

no code implementations • 1 Jan 2021 • Gregory Bonaert, Youri Coppens, Denis Steckelmacher, Ann Nowe

Our key contribution to improve explainability is introducing goal-based explanations, a new explanation mechanism where the agent produces goals and attempts to reach those goals one-by-one while maximizing the collected reward.

reinforcement-learning Reinforcement Learning (RL)