Docs

Notes and write-ups.

• VAE — from the problem up
  Deriving the Variational Autoencoder from first principles.
• VAE — the historical path
  How VAEs emerged, traced through their historical development.
• Reward Models — for robot learning
  Reward models in the context of robot learning.
• Reward Models — Survey
  Survey of reward modeling approaches.
• GR00T Policy Inference — Hierarchical Control
  How a trained GR00T manipulation policy runs on the real Unitree G1: chunked policy, two clocks, WBC stack.