条件付きフローマッチングによる効率的なマルチタスクマニピュレーションポリシーの蒸留
Distillation for Efficient Multitask Manipulation Policies via Conditional Flow Matching
単一タスクの条件付きフローマッチング専門家から速度場を蒸留し、モデルサイズを固定したままマルチタスク操作ポリシーの性能を向上させる手法を提案。
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: Shreya Deshmukh, Imen Mahdi, Nick Heppert, Abhinav Valada
分類: cs.RO, cs.AI, cs.LG
原文アブストラクト
Advances in generative modeling have recently been extensively employed in robotics for policy learning. In particular, Conditional Flow Matching (CFM) trained with expert demonstrations has been shown to outperform existing methods on robot manipulation benchmarks. While prior work has mainly focused on single-task settings, we study the problem from a multi-task perspective, as training independent models for each task is computationally expensive. Multi-Task policy learning comes with its own set of challenges, as naively training on a concatenated dataset of demonstrations would either require increased model capacity to accommodate the added complexity or result in drops in performance. We propose to distill knowledge from single-task CFM experts into a shared multi-task policy by transferring their learned velocity fields. We combine this distillation signal with the original CFM objective to retain fidelity to the demonstrations. Experiments on RLBench show that our approach improves multi-task policy performance over naive training while maintaining a fixed model size.