IN-RIL: 模倣学習と強化学習を交互に組み合わせた方策ファインチューニング
IN-RIL: Interleaved Reinforcement and Imitation Learning for Policy Fine-Tuning
強化学習の更新の合間に模倣学習の更新を定期的に挿入し、勾配を直交部分空間に分離することで、ファインチューニングの安定性とサンプル効率を向上させる手法を提案した。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Dechen Gao, Hang Wang, Hanchu Zhou, Nejib Ammar, Shatadal Mishra, Ahmadreza Moradipari, Iman Soltani, Junshan Zhang
分類: cs.RO, cs.AI
原文アブストラクト
Imitation learning (IL) and reinforcement learning (RL) each offer distinct advantages for robotics policy learning: IL provides stable learning from demonstrations, and RL promotes generalization through exploration. While existing robot learning approaches using IL-based pre-training followed by RL-based fine-tuning are promising, this two-step learning paradigm often suffers from instability and poor sample efficiency during the RL fine-tuning phase. In this work, we introduce IN-RIL, INterleaved Reinforcement learning and Imitation Learning, for policy fine-tuning, which periodically injects IL updates after multiple RL updates and hence can benefit from the stability of IL and the guidance of expert data for more efficient exploration throughout the entire fine-tuning process. Since IL and RL involve different optimization objectives, we develop gradient separation mechanisms to prevent destructive interference during \ABBR fine-tuning, by separating possibly conflicting gradient updates in orthogonal subspaces. Furthermore, we conduct rigorous analysis, and our findings shed light on why interleaving IL with RL stabilizes learning and improves sample-efficiency. Extensive experiments on 14 robot manipulation and locomotion tasks across 3 benchmarks, including FurnitureBench, OpenAI Gym, and Robomimic, demonstrate that \ABBR can significantly improve sample efficiency and mitigate performance collapse during online finetuning in both long- and short-horizon tasks with either sparse or dense rewards. IN-RIL, as a general plug-in compatible with various state-of-the-art RL algorithms, can significantly improve RL fine-tuning, e.g., from 12\% to 88\% with 6.3x improvement in the success rate on Robomimic Transport. Project page: https://github.com/ucd-dare/IN-RIL.
関連論文
- 高次行動監督による強力な方策クラス模倣学習/強化学習
- ConTrack: 適応的トレードオフ制御による拘束付き手の動作追跡模倣学習/強化学習
- 不完全なデモンストレーションからの学習:時間行動木ガイドによる軌道修復模倣学習/強化学習
- DFM: 表現豊かなダンス動作学習のための深層フーリエ模倣模倣学習/強化学習
- BMP:Bスプラインと運動プリミティブの橋渡し模倣学習/強化学習
- 強化学習によるリザバーダイナミクス変調を用いた効率的なロボットスキル合成模倣学習/強化学習
