CARF: 失敗誘導フローマッチングの対比的引力-斥力
CARF: Contrastive Attraction-Repulsion of Failure-Guided Flow Matching
失敗軌道から「進展する部分」は模倣し「失敗を招く部分」は回避する非対称な学習を、フローマッチングで統一的に実現する手法を提案。
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: Shuqi Zhao, Bang Du, Cheng-En Wu, Yichen Xie, Yixiao Wang, Masayoshi Tomizuka
分類: cs.RO
原文アブストラクト
Robot demonstration collection often produces imperfect or failed trajectories in addition to successful demonstrations. Existing methods typically exploit failed trajectories by identifying segments that still make progress toward task completion, but largely overlook \textit{failure-critical behaviors} that directly lead to task failure. Here we argue that these two types of segments provide fundamentally asymmetric supervision: progressive segments should be imitated, whereas failure-critical segments should be explicitly avoided. Based on this observation, we propose CARF, a Contrastive Attraction-Repulsion of Failure-guided framework for learning from imperfect robot data. CARF introduces a progress-based importance scorer, trained solely on successful expert demonstrations and its perturbation results, to estimate step-wise contributions toward task completion and identify informative regions in failed trajectories. These scores guide a unified flow-matching objective that attracts the policy toward progressive behaviors and repels it from failure-critical ones, while excluding ambiguous segments. This enables more comprehensive utilization of imperfect data and avoids unreliable supervision from ambiguous failure segments. Extensive experiments in simulation and the real world demonstrate consistent improvements over competing baselines across diverse failure scenarios, with ablations further validating the effectiveness of the proposed scoring and attraction-repulsion mechanisms. Our website is https://zhao-sq.github.io/carf/#.