Many recent robot learning papers, such as Hi-robot, Gemini-Robotics, etc., use different types of hierarchical planning. In this lecture, I review the background in hierarchical planning and reinforcement learning on a case of humanoid control, before getting into the details of how these methods are used in recent papers. These topics include exploration benefits, problem decomposition, credit assignment, challenges in training a good low-level policy, and co-training policies together.
Continue this lesson in the app
Install CourseHive on Android or iOS to keep learning while you move.
