汎用ロボット方策におけるショートカット学習:データセットの多様性と断片化の役割
Shortcut Learning in Generalist Robot Policies: The Role of Dataset Diversity and Fragmentation
大規模ロボットデータセットで訓練された汎用方策が訓練分布外で汎化しにくい原因を分析し、タスク無関係な特徴に依存する「ショートカット学習」が主因であること、サブデータセットの多様性不足と分布の断片化がそれを助長することを示した。さらにデータ拡張でショートカット学習を軽減し汎化性能を改善できることを実証した。
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: Youguang Xing, Xu Luo, Junlin Xie, Lianli Gao, Hengtao Shen, Jingkuan Song
分類: cs.RO, cs.AI, cs.CV
原文アブストラクト
Generalist robot policies trained on large-scale datasets such as Open X-Embodiment (OXE) demonstrate strong performance across a wide range of tasks. However, they often struggle to generalize beyond the distribution of their training data. In this paper, we investigate the underlying cause of this limited generalization capability. We identify shortcut learning -- the reliance on task-irrelevant features -- as a key impediment to generalization. Through comprehensive theoretical and empirical analysis, we uncover two primary contributors to shortcut learning: (1) limited diversity within individual sub-datasets, and (2) significant distributional disparities across sub-datasets, leading to dataset fragmentation. These issues arise from the inherent structure of large-scale datasets like OXE, which are typically composed of multiple sub-datasets collected independently across varied environments and embodiments. Our findings provide critical insights into dataset collection strategies that can reduce shortcut learning and enhance the generalization ability of generalist robot policies. Moreover, in scenarios where acquiring new large-scale data is impractical, we demonstrate that carefully selected robotic data augmentation strategies can effectively reduce shortcut learning in existing offline datasets, thereby improving generalization capabilities of generalist robot policies, e.g., $\pi_0$, in both simulation and real-world environments. More information at https://lucky-light-sun.github.io/proj/shortcut-learning-in-grps/.
