DiffEM: Learning from Corrupted Data with Diffusion Models via Expectation Maximization

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hosseintabar, Danial, Chen, Fan, Daras, Giannis, Torralba, Antonio, Daskalakis, Constantinos
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918256575512576
author Hosseintabar, Danial
Chen, Fan
Daras, Giannis
Torralba, Antonio
Daskalakis, Constantinos
author_facet Hosseintabar, Danial
Chen, Fan
Daras, Giannis
Torralba, Antonio
Daskalakis, Constantinos
contents Diffusion models have emerged as powerful generative priors for high-dimensional inverse problems, yet learning them when only corrupted or noisy observations are available remains challenging. In this work, we propose a new method for training diffusion models with Expectation-Maximization (EM) from corrupted data. Our proposed method, DiffEM, utilizes conditional diffusion models to reconstruct clean data from observations in the E-step, and then uses the reconstructed data to refine the conditional diffusion model in the M-step. Theoretically, we provide monotonic convergence guarantees for the DiffEM iteration, assuming appropriate statistical conditions. We demonstrate the effectiveness of our approach through experiments on various image reconstruction tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2510_12691
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DiffEM: Learning from Corrupted Data with Diffusion Models via Expectation Maximization
Hosseintabar, Danial
Chen, Fan
Daras, Giannis
Torralba, Antonio
Daskalakis, Constantinos
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Diffusion models have emerged as powerful generative priors for high-dimensional inverse problems, yet learning them when only corrupted or noisy observations are available remains challenging. In this work, we propose a new method for training diffusion models with Expectation-Maximization (EM) from corrupted data. Our proposed method, DiffEM, utilizes conditional diffusion models to reconstruct clean data from observations in the E-step, and then uses the reconstructed data to refine the conditional diffusion model in the M-step. Theoretically, we provide monotonic convergence guarantees for the DiffEM iteration, assuming appropriate statistical conditions. We demonstrate the effectiveness of our approach through experiments on various image reconstruction tasks.
title DiffEM: Learning from Corrupted Data with Diffusion Models via Expectation Maximization
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.12691