Learning safe, constrained policies via imitation learning: Connection to Probabilistic Inference and a Naive Algorithm

Open in new window