Hierarchical Policy

First saw this from Hierarchical Behavior Cloning from Robomimic paper.

“HBC consists of a low-level policy that is conditioned on future observations sg ∈ S (termed subgoals) and outputs action sequences to try and achieve them, and a high-level policy that predicts future subgoals from the current observation.”

• https://arxiv.org/pdf/2108.03298

So we have two policies:

• ${\pi }_{\theta }^H(s)$ that predicts $s_{t+T}$ (implemented as
• ${\pi }_{\theta }^L(s,s_g)$ that predicts $a$ to get to the goal states

• https://www.figure.ai/news/helix