- document
-
Coppens, Youri (author), Steckelmacher, Denis (author), Jonker, C.M. (author), Nowe, A.S.P. (author)Today’s advanced Reinforcement Learning algorithms produce black-box policies, that are often difficult to interpret and trust for a person. We introduce a policy distilling algorithm, building on the CN2 rule mining algorithm, that distills the policy into a rule-based decision system. At the core of our approach is the fact that an RL...conference paper 2021