Reproduce “AtomicRL: Atomic Reinforcement Learning for Long-Horizon Robotic Manipulation” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Likui Zhang, Zhe Zhu, Zhihao Zhan, Yibo Bai, Hanlin Wang, Dongwei Li, Bin Wang, Tianle Zhang and 2 more
AtomicRL uses semantic subtask boundaries to limit how generalized advantage estimation carries learning signals backward after failures, while preserving within-subtask learning and directing exploration toward weak steps. On held-out CALVIN ABC-D, it raises five-task completion from PPO’s 70.6% to 80.1%. On AgiBot G1 tabletop tasks, it averages 69.3% versus SFT’s 48.0% after simulation-based training; results cover specified evaluations.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.