Diffusion Model for Manifold Data: Score Decomposition, Curvature, and Statistical Complexity
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910171703279616 |
|---|---|
| author | Zhang, Zixuan Huang, Kaixuan Zhao, Tuo Wang, Mengdi Chen, Minshuo |
| author_facet | Zhang, Zixuan Huang, Kaixuan Zhao, Tuo Wang, Mengdi Chen, Minshuo |
| contents | Diffusion models have become a leading framework in generative modeling, yet their theoretical understanding -- especially for high-dimensional data concentrated on low-dimensional structures -- remains incomplete. This paper investigates how diffusion models learn such structured data, focusing on two key aspects: statistical complexity and influence of data geometric properties. By modeling data as samples from a smooth Riemannian manifold, our analysis reveals crucial decompositions of score functions in diffusion models under different levels of injected noise. We also highlight the interplay of manifold curvature with the structures in the score function. These analyses enable an efficient neural network approximation to the score function, built upon which we further provide statistical rates for score estimation and distribution learning. Remarkably, the obtained statistical rates are governed by the intrinsic dimension of data and the manifold curvature. These results advance the statistical foundations of diffusion models, bridging theory and practice for generative modeling on manifolds. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_20645 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Diffusion Model for Manifold Data: Score Decomposition, Curvature, and Statistical Complexity Zhang, Zixuan Huang, Kaixuan Zhao, Tuo Wang, Mengdi Chen, Minshuo Machine Learning Diffusion models have become a leading framework in generative modeling, yet their theoretical understanding -- especially for high-dimensional data concentrated on low-dimensional structures -- remains incomplete. This paper investigates how diffusion models learn such structured data, focusing on two key aspects: statistical complexity and influence of data geometric properties. By modeling data as samples from a smooth Riemannian manifold, our analysis reveals crucial decompositions of score functions in diffusion models under different levels of injected noise. We also highlight the interplay of manifold curvature with the structures in the score function. These analyses enable an efficient neural network approximation to the score function, built upon which we further provide statistical rates for score estimation and distribution learning. Remarkably, the obtained statistical rates are governed by the intrinsic dimension of data and the manifold curvature. These results advance the statistical foundations of diffusion models, bridging theory and practice for generative modeling on manifolds. |
| title | Diffusion Model for Manifold Data: Score Decomposition, Curvature, and Statistical Complexity |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2603.20645 |