EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Operator Analysis on Stochastic Differential Equation (SDE)-Based Diffusion Generative Models

Yunpei Wu iD, Yoshinobu Kawahara iD

Entropy · 04.03.2026 · 28 (3) · 290

DOI: 10.3390/e28030290 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Entropy, online erschienen am 04.03.2026. Verfasst von Yunpei Wu und 1 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Mathematics, Kyushu University, Fukuoka 819-0395, Japan. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Score-based generative models, grounded in stochastic differential equations (SDEs), excel in producing high-quality data but suffer from slow sampling due to the extensive nonlinear computations required for iterative score function evaluations. We propose an innovative approach that integrates score-based reverse SDEs with kernel methods, leveraging the derivative reproducing property of reproducing kernel Hilbert spaces (RKHSs) to efficiently approximate the eigenfunctions and eigenvalues of the Fokker–Planck operator. This enables data generation through linear combinations of eigenfunctions, transforming computationally intensive nonlinear operations into efficient linear ones, thereby significantly reducing computational overhead. Notably, our experimental results demonstrate remarkable progress: despite a slight reduction in sample diversity, the sampling time for a single image on the CIFAR-10 dataset is reduced to an impressive 0.29 s, marking a substantial advancement in efficiency. This work introduces novel theoretical and practical tools for generative modeling, establishing a robust foundation for real-time applications.

Schlagwörter

Stichwörter der Autor:innen

Fokker–Planck operatorSDEeigenfunction decompositiongenerative modelingkernel methods

Autor:innen und Institutionen

  • Faculty of Mathematics, Kyushu University, Fukuoka 819-0395, Japan.
  • Graduate School of Information Science and Technology, The University of Osaka, Osaka 565-0871, Japan.

ORCID-Profile: Yunpei Wu, Yoshinobu Kawahara

Zitieren

APA

Wu, Y., & Kawahara, Y. (2026). An Operator Analysis on Stochastic Differential Equation (SDE)-Based Diffusion Generative Models. Entropy, 28(3), 290. https://doi.org/10.3390/e28030290

Vancouver

Wu Y, Kawahara Y. An Operator Analysis on Stochastic Differential Equation (SDE)-Based Diffusion Generative Models. Entropy. 2026;28(3):290. doi:10.3390/e28030290

BibTeX
@article{Wu2026,
  author = {Wu, Yunpei and Kawahara, Yoshinobu},
  title = {An Operator Analysis on Stochastic Differential Equation (SDE)-Based Diffusion Generative Models},
  journal = {Entropy},
  year = {2026},
  volume = {28},
  number = {3},
  pages = {290},
  doi = {10.3390/e28030290}
}
Kostenlos starten →