Relevant:

Comment:

Leveraging the human knowledge and feedback for reinforcement learning (human-in-the-loop)