Explainable Reinforcement Learning Through a Causal Lens
This work addresses the need for better interpretability in AI systems, particularly for users interacting with reinforcement learning agents, though it is incremental by applying existing causal methods to a new domain.
The paper tackles the problem of explaining reinforcement learning agent behavior by using causal models to generate explanations based on counterfactual analysis, and it shows that these explanations improve participant understanding, satisfaction, and trust compared to baselines in a study with 120 participants.
Prevalent theories in cognitive science propose that humans understand and represent the knowledge of the world through causal relationships. In making sense of the world, we build causal models in our mind to encode cause-effect relations of events and use these to explain why new events happen. In this paper, we use causal models to derive causal explanations of behaviour of reinforcement learning agents. We present an approach that learns a structural causal model during reinforcement learning and encodes causal relationships between variables of interest. This model is then used to generate explanations of behaviour based on counterfactual analysis of the causal model. We report on a study with 120 participants who observe agents playing a real-time strategy game (Starcraft II) and then receive explanations of the agents' behaviour. We investigated: 1) participants' understanding gained by explanations through task prediction; 2) explanation satisfaction and 3) trust. Our results show that causal model explanations perform better on these measures compared to two other baseline explanation models.