Offline reinforcement learning

This note has no content.