2606.12365v1 Jun 10, 2026 cs.RO

Ambient Diffusion Policy: Imitation Learning from Suboptimal Data in Robotics

G. Daras
G. Daras
Citations: 2,849
h-index: 18
Nicholas Pfaff
Nicholas Pfaff
MIT
Citations: 77
h-index: 4
Russ Tedrake
Russ Tedrake
Citations: 1,195
h-index: 14
Constantinos Daskalakis
Constantinos Daskalakis
Citations: 240
h-index: 10
Thomas Cohn
Thomas Cohn
MIT
Citations: 86
h-index: 5
Adam Wei
Adam Wei
Citations: 72
h-index: 2
A. K. Dayi
A. K. Dayi
Citations: 43
h-index: 4

We propose Ambient Diffusion Policy, a simple and principled method for imitation learning from suboptimal data in robotics. High-quality, task-specific robot data is expensive and time-consuming to collect, while suboptimal datasets with lower-quality or out-of-distribution demonstrations are abundant. Existing methods that co-train on both data sources in robotics often fail to separate the meaningful and the harmful features in the suboptimal samples. In contrast, our method extracts only the useful features by introducing a new axis to co-training in robotics: noise-dependent data usage. Ambient Diffusion Policy restricts the contribution of suboptimal data during training to only the high and low diffusion times. To rigorously justify our approach, we first observe that robot action data exhibits a spectral power law. This induces two important properties on the optimal Diffusion Policy that we exploit: a global-to-local hierarchy and locality. We theoretically formalize this discussion using a simplified model. Our experiments validate Ambient Diffusion Policy on four types of suboptimal action data (noisy trajectories, sim-to-real gap, task mismatch, and large-scale data mixtures) across six tasks. The results show that it effectively learns from arbitrary sources of suboptimal data. Notably, it outperforms existing co-training baselines by up to 33% when scaled to Open X-Embodiment - a large dataset with heterogeneous data quality and unstructured distribution shifts. Overall, Ambient Diffusion Policy increases the utility of suboptimal demonstrations and expands the set of usable data sources in robotics.

0 Citations
0 Influential
9 Altmetric
45.0 Score
Original PDF

No Analysis Report Yet

This paper hasn't been analyzed by Gemini yet.

Log in to request an AI analysis.

댓글

댓글을 작성하려면 로그인하세요.

아직 댓글이 없습니다. 첫 번째 댓글을 남겨보세요!