Diffusion models for Gaussian distributions: Exact solutions and Wasserstein errors - Institut Denis Poisson Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2024

Diffusion models for Gaussian distributions: Exact solutions and Wasserstein errors

Emile Pierret
  • Fonction : Auteur
  • PersonId : 1384593

Résumé

Diffusion or score-based models recently showed high performance in image generation. They rely on a forward and a backward stochastic differential equations (SDE). The sampling of a data distribution is achieved by solving numerically the backward SDE or its associated flow ODE. Studying the convergence of these models necessitates to control four different types of error: the initialization error, the truncation error, the discretization and the score approximation. In this paper, we study theoretically the behavior of diffusion models and their numerical implementation when the data distribution is Gaussian. In this restricted framework where the score function is a linear operator, we can derive the analytical solutions of the forward and backward SDEs as well as the associated flow ODE. This provides exact expressions for various Wasserstein errors which enable us to compare the influence of each error type for any sampling scheme, thus allowing to monitor convergence directly in the data space instead of relying on Inception features. Our experiments show that the recommended numerical schemes from the diffusion models literature are also the best sampling schemes for Gaussian distributions.
Fichier principal
Vignette du fichier
Gaussian_diffusion_W2.pdf (3.45 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04584209 , version 1 (23-05-2024)
hal-04584209 , version 2 (07-06-2024)
hal-04584209 , version 3 (12-06-2024)

Identifiants

  • HAL Id : hal-04584209 , version 1

Citer

Emile Pierret, Bruno Galerne. Diffusion models for Gaussian distributions: Exact solutions and Wasserstein errors. 2024. ⟨hal-04584209v1⟩
0 Consultations
0 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More