Immiscible Diffusion Policy: Preserving Multimodal Robot Actions through Label-Free Noise Assignment
To alleviate this problem, we propose Immiscible Diffusion Policy, a label-free training-time add-on to diffusion policy that uses action-noise assignment to preserve relatively distinct noise-to-action routes without modifying the policy architecture or inference procedure.
ProofPaper ↗
Key points
- When diffusion policies were first introduced, they were expected to recover multi-modal action distributions.
- However, we find this expectation does not always hold, as diffusion policies often collapse to a single modality even when we guarantee the balance of dataset modalities and exact within-batch symmetry.
- Our analysis indicates that independent action-noise pairing contributes to this failure by increasing mixing and crossing among diffusion paths, which can produce averaged denoising responses and suppress modality-specific behavior.
- These results demonstrate that Immiscible Diffusion Policy provides a simple yet robust approach to preserving action multi-modality in general robot learning tasks.
Sources (1)
- [1]Immiscible Diffusion Policy: Preserving Multimodal Robot Actions through Label-Free Noise AssignmentarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 03:22 AM
To alleviate this problem, we propose Immiscible Diffusion Policy, a label-free training-time add-on to diffusion policy that uses action-noise assignment to preserve relatively distinct noise-to-action routes without modifying the policy architecture or inference procedure.
When diffusion policies were first introduced, they were expected to recover multi-modal action distributions.
Extractive summary: sentences quoted from the sources.