Generalization error bound for denoising score matching under relaxed manifold assumption

2025/02/19 by Konstantin Yakovlev, Yakovlev, Konstantin, Nikita Puchkin +1 · 7 citations
Computer Science · #Face and Expression Recognition

paper · pdf · doi:10.48550/arxiv.2502.13662

Abstract

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.

Cited by

• Simultaneous Approximation of the Score Function and Its Derivatives by Deep Neural Networks
• Approximation Capabilities of Feedforward Neural Networks with GELU Activations
• Rao-Blackwellized Score Matching on Manifolds
• Distribution estimation via Flow Matching with Lipschitz guarantees
• Generalization bounds for score-based generative models: a synthetic proof
• Localized Diffusion Models
• On the minimax optimality of Flow Matching through the connection to kernel density estimation

Related

• Rao-Blackwellized Score Matching on Manifolds
• Optimal Denoising in Score-Based Generative Models: The Role of Data Regularity
• Target Score Matching
• Heavy-tailed denoising score matching
• A Connection Between Score Matching and Local Intrinsic Dimension