Explainability in Deep Reinforcement Learning, a Review into Current Methods and Applications
It tackles the problem of trust and understanding in DRL for researchers and the public, but it is incremental as it reviews existing methods without introducing new ones.
This review addresses the lack of interpretability in Deep Reinforcement Learning (DRL) by surveying current Explainable AI (XAI) methods, such as Decision Trees and Shapley Values, to identify the best-suited models for various applications and highlight potential underutilized approaches.
The use of Deep Reinforcement Learning (DRL) schemes has increased dramatically since their first introduction in 2015. Though uses in many different applications are being found, they still have a problem with the lack of interpretability. This has bread a lack of understanding and trust in the use of DRL solutions from researchers and the general public. To solve this problem, the field of Explainable Artificial Intelligence (XAI) has emerged. This entails a variety of different methods that look to open the DRL black boxes, ranging from the use of interpretable symbolic Decision Trees (DT) to numerical methods like Shapley Values. This review looks at which methods are being used and for which applications. This is done to identify which models are the best suited to each application or if a method is being underutilised.