Implicit score matching meets denoising score matching: improved rates of convergence and log-density Hessian estimation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yakovlev, Konstantin, Markovich, Anna, Puchkin, Nikita
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912797292494848
author Yakovlev, Konstantin
Markovich, Anna
Puchkin, Nikita
author_facet Yakovlev, Konstantin
Markovich, Anna
Puchkin, Nikita
contents We study the problem of estimating the score function using both implicit score matching and denoising score matching. Assuming that the data distribution exhibiting a low-dimensional structure, we prove that implicit score matching is able not only to adapt to the intrinsic dimension, but also to achieve the same rates of convergence as denoising score matching in terms of the sample size. Furthermore, we demonstrate that both methods allow us to estimate log-density Hessians without the curse of dimensionality by simple differentiation. This justifies convergence of ODE-based samplers for generative diffusion models. Our approach is based on Gagliardo-Nirenberg-type inequalities relating weighted $L^2$-norms of smooth functions and their derivatives.
format Preprint
id arxiv_https___arxiv_org_abs_2512_24378
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Implicit score matching meets denoising score matching: improved rates of convergence and log-density Hessian estimation
Yakovlev, Konstantin
Markovich, Anna
Puchkin, Nikita
Statistics Theory
Machine Learning
We study the problem of estimating the score function using both implicit score matching and denoising score matching. Assuming that the data distribution exhibiting a low-dimensional structure, we prove that implicit score matching is able not only to adapt to the intrinsic dimension, but also to achieve the same rates of convergence as denoising score matching in terms of the sample size. Furthermore, we demonstrate that both methods allow us to estimate log-density Hessians without the curse of dimensionality by simple differentiation. This justifies convergence of ODE-based samplers for generative diffusion models. Our approach is based on Gagliardo-Nirenberg-type inequalities relating weighted $L^2$-norms of smooth functions and their derivatives.
title Implicit score matching meets denoising score matching: improved rates of convergence and log-density Hessian estimation
topic Statistics Theory
Machine Learning
url https://arxiv.org/abs/2512.24378