Semifactual Explanations for Reinforcement Learning
This work addresses the need for interpretability in DRL to enhance user trust and integration in real-life tasks, representing an incremental advancement by applying semifactual explanations, previously used in psychology and supervised learning, to RL.
The paper tackles the problem of explaining deep reinforcement learning (DRL) agent decisions by introducing the first approach to generating semifactual explanations, which provide 'even if' scenarios, and finds that their algorithms produce explanations that are easier to reach, better represent the policy, and are more diverse compared to baselines in two standard RL environments.
Reinforcement Learning (RL) is a learning paradigm in which the agent learns from its environment through trial and error. Deep reinforcement learning (DRL) algorithms represent the agent's policies using neural networks, making their decisions difficult to interpret. Explaining the behaviour of DRL agents is necessary to advance user trust, increase engagement, and facilitate integration with real-life tasks. Semifactual explanations aim to explain an outcome by providing "even if" scenarios, such as "even if the car were moving twice as slowly, it would still have to swerve to avoid crashing". Semifactuals help users understand the effects of different factors on the outcome and support the optimisation of resources. While extensively studied in psychology and even utilised in supervised learning, semifactuals have not been used to explain the decisions of RL systems. In this work, we develop a first approach to generating semifactual explanations for RL agents. We start by defining five properties of desirable semifactual explanations in RL and then introducing SGRL-Rewind and SGRL-Advance, the first algorithms for generating semifactual explanations in RL. We evaluate the algorithms in two standard RL environments and find that they generate semifactuals that are easier to reach, represent the agent's policy better, and are more diverse compared to baselines. Lastly, we conduct and analyse a user study to assess the participant's perception of semifactual explanations of the agent's actions.