CoDA: Co-Evolution of Data Engines and Action Policies
[Robotyka] Autorzy: Tianjiao Yu, Zhilin Ou, Yifan Shen i 5 więcej
Generative simulation can automate robot environment construction, but physically valid environments are not necessarily useful for training. The challenge is to identify environments that address a policy's failures and supervision that helps it overcome them. We introduce CoDA, a framework in which a simulation data engine and an action policy co-evolve through a shared heterogeneous memory of environment construction, policy failures, and learning progress. The memory compiles validated diagnoses from a vision-language critic into generation programs that recreate difficulties and training programs that help the policy overcome them. We further propose Text2potential to convert these diagnoses into staged potentials for action-chunk value learning, providing dense supervision for policy updates while preserving the task's optimal policy under standard shaping assumptions. Measured le…
- Kategoria arxiv
- cs.RO
- Kategoria
- Robotyka