Designing Interpretable Approximations to Deep Reinforcement Learning with Soft Decision Trees

Open in new window