Saved in:
Bibliographic Details
Main Authors: Leong, Oscar, Traonmilin, Yann
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.20511
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918147460694016
author Leong, Oscar
Traonmilin, Yann
author_facet Leong, Oscar
Traonmilin, Yann
contents Recovering high-dimensional signals from corrupted measurements is a central challenge in inverse problems. Recent advances in generative diffusion models have shown remarkable empirical success in providing strong data-driven priors, but rigorous recovery guarantees remain limited. In this work, we develop a theoretical framework for analyzing deterministic diffusion-based algorithms for inverse problems, focusing on a deterministic version of the algorithm proposed by Kadkhodaie \& Simoncelli \cite{kadkhodaie2021stochastic}. First, we show that when the underlying data distribution concentrates on a low-dimensional model set, the associated noise-convolved scores can be interpreted as time-varying projections onto such a set. This leads to interpreting previous algorithms using diffusion priors for inverse problems as generalized projected gradient descent methods with varying projections. When the sensing matrix satisfies a restricted isometry property over the model set, we can derive quantitative convergence rates that depend explicitly on the noise schedule. We apply our framework to two instructive data distributions: uniform distributions over low-dimensional compact, convex sets and low-rank Gaussian mixture models. In the latter setting, we can establish global convergence guarantees despite the nonconvexity of the underlying model set.
format Preprint
id arxiv_https___arxiv_org_abs_2509_20511
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Recovery Theory for Diffusion Priors: Deterministic Analysis of the Implicit Prior Algorithm
Leong, Oscar
Traonmilin, Yann
Machine Learning
Signal Processing
Optimization and Control
Recovering high-dimensional signals from corrupted measurements is a central challenge in inverse problems. Recent advances in generative diffusion models have shown remarkable empirical success in providing strong data-driven priors, but rigorous recovery guarantees remain limited. In this work, we develop a theoretical framework for analyzing deterministic diffusion-based algorithms for inverse problems, focusing on a deterministic version of the algorithm proposed by Kadkhodaie \& Simoncelli \cite{kadkhodaie2021stochastic}. First, we show that when the underlying data distribution concentrates on a low-dimensional model set, the associated noise-convolved scores can be interpreted as time-varying projections onto such a set. This leads to interpreting previous algorithms using diffusion priors for inverse problems as generalized projected gradient descent methods with varying projections. When the sensing matrix satisfies a restricted isometry property over the model set, we can derive quantitative convergence rates that depend explicitly on the noise schedule. We apply our framework to two instructive data distributions: uniform distributions over low-dimensional compact, convex sets and low-rank Gaussian mixture models. In the latter setting, we can establish global convergence guarantees despite the nonconvexity of the underlying model set.
title A Recovery Theory for Diffusion Priors: Deterministic Analysis of the Implicit Prior Algorithm
topic Machine Learning
Signal Processing
Optimization and Control
url https://arxiv.org/abs/2509.20511