On the Regularity and Generalization of One-Step Wasserstein-guided Generative Models for PDE-Induced Measures
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916031910379520 |
|---|---|
| author | Lin, Likun Wang, Zhongjian Xin, Jack Zhang, Zhiwen |
| author_facet | Lin, Likun Wang, Zhongjian Xin, Jack Zhang, Zhiwen |
| contents | Despite the remarkable empirical success of generative models, the available theory on their statistical accuracy in scientific computing remains largely pessimistic. This paper develops a theoretical framework for understanding the regularity of transport maps and the generalization properties of one-step Wasserstein-guided generative models for PDE-induced probability measures. We consider normalized target densities associated with linear elliptic and parabolic equations on bounded domains, as well as diffusion and Fokker--Planck equations on the torus. Under standard structural assumptions, we prove that these target measures satisfy doubling conditions. By combining this fact with regularity theory for optimal transport between doubling measures, we show that the optimal transport map from a uniform source measure to the target measure is Hölder continuous. This regularity yields an approximation-theoretic justification for one-step generative models that learn PDE-induced distributions via a single pushforward map. As a representative instance, we study DeepParticle and derive excess-risk bounds characterizing the discrepancy between the learned map and the population-optimal map. We also establish a robustness estimate under target shift and illustrate the theory with experiments which support the derived rates. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_21388 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | On the Regularity and Generalization of One-Step Wasserstein-guided Generative Models for PDE-Induced Measures Lin, Likun Wang, Zhongjian Xin, Jack Zhang, Zhiwen Machine Learning Artificial Intelligence Numerical Analysis 68T07, 35J25, 41A46, 35K20, 35Q84 Despite the remarkable empirical success of generative models, the available theory on their statistical accuracy in scientific computing remains largely pessimistic. This paper develops a theoretical framework for understanding the regularity of transport maps and the generalization properties of one-step Wasserstein-guided generative models for PDE-induced probability measures. We consider normalized target densities associated with linear elliptic and parabolic equations on bounded domains, as well as diffusion and Fokker--Planck equations on the torus. Under standard structural assumptions, we prove that these target measures satisfy doubling conditions. By combining this fact with regularity theory for optimal transport between doubling measures, we show that the optimal transport map from a uniform source measure to the target measure is Hölder continuous. This regularity yields an approximation-theoretic justification for one-step generative models that learn PDE-induced distributions via a single pushforward map. As a representative instance, we study DeepParticle and derive excess-risk bounds characterizing the discrepancy between the learned map and the population-optimal map. We also establish a robustness estimate under target shift and illustrate the theory with experiments which support the derived rates. |
| title | On the Regularity and Generalization of One-Step Wasserstein-guided Generative Models for PDE-Induced Measures |
| topic | Machine Learning Artificial Intelligence Numerical Analysis 68T07, 35J25, 41A46, 35K20, 35Q84 |
| url | https://arxiv.org/abs/2605.21388 |