データの類推が効率的な異種ロボット間転移を可能にする
Data Analogies Enable Efficient Cross-Embodiment Transfer
異なるロボット間での模倣学習の転移において、単にデータ量を増やすよりも、シーンやタスクを揃えたペアデータ（データ類推）が形態の違いに対して効果的であることを示し、実世界で成功率を平均22.5%向上させた。
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: Jonathan Yang, Chelsea Finn, Dorsa Sadigh
分類: cs.RO
原文アブストラクト
Generalist robot policies are trained on demonstrations collected across a wide variety of robots, scenes, and viewpoints. Yet it remains unclear how to best organize and scale such heterogeneous data so that it genuinely improves performance in a given target setting. In this work, we ask: what form of demonstration data is most useful for enabling transfer across robot set-ups? We conduct controlled experiments that vary end-effector morphology, robot platform appearance, and camera perspective, and compare the effects of simply scaling the number of demonstrations against systematically broadening the diversity in different ways. Our simulated experiments show that while perceptual shifts such as viewpoint benefit most from broad diversity, morphology shifts benefit far less from unstructured diversity and instead see the largest gains from data analogies, i.e. paired demonstrations that align scenes, tasks, and/or trajectories across different embodiments. Informed by the simulation results, we improve real-world cross-embodiment transfer success by an average of $22.5\%$ over large-scale, unpaired datasets by changing only the composition of the data.
