Generalization error bound for denoising score matching under relaxed manifold assumption

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yakovlev, Konstantin, Puchkin, Nikita
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914225848320000
author Yakovlev, Konstantin
Puchkin, Nikita
author_facet Yakovlev, Konstantin
Puchkin, Nikita
contents We examine theoretical properties of the denoising score matching estimate. We model the density of observations with a nonparametric Gaussian mixture. We significantly relax the standard manifold assumption allowing the samples step away from the manifold. At the same time, we are still able to leverage a nice distribution structure. We derive non-asymptotic bounds on the approximation and generalization errors of the denoising score matching estimate. The rates of convergence are determined by the intrinsic dimension. Furthermore, our bounds remain valid even if we allow the ambient dimension grow polynomially with the sample size.
format Preprint
id arxiv_https___arxiv_org_abs_2502_13662
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Generalization error bound for denoising score matching under relaxed manifold assumption
Yakovlev, Konstantin
Puchkin, Nikita
Machine Learning
Statistics Theory
We examine theoretical properties of the denoising score matching estimate. We model the density of observations with a nonparametric Gaussian mixture. We significantly relax the standard manifold assumption allowing the samples step away from the manifold. At the same time, we are still able to leverage a nice distribution structure. We derive non-asymptotic bounds on the approximation and generalization errors of the denoising score matching estimate. The rates of convergence are determined by the intrinsic dimension. Furthermore, our bounds remain valid even if we allow the ambient dimension grow polynomially with the sample size.
title Generalization error bound for denoising score matching under relaxed manifold assumption
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2502.13662