Learning Normalized Energy Models for Linear Inverse Problems

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zilberstein, Nicolas, Segarra, Santiago, Simoncelli, Eero, Guth, Florentin
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909051731836928
author Zilberstein, Nicolas
Segarra, Santiago
Simoncelli, Eero
Guth, Florentin
author_facet Zilberstein, Nicolas
Segarra, Santiago
Simoncelli, Eero
Guth, Florentin
contents Generative diffusion models can provide powerful prior probability models for inverse problems in imaging, but existing implementations suffer from two key limitations: $(i)$ the prior density is represented implicitly, and $(ii)$ they rely on likelihood approximations that introduce sampling biases. We address these challenges by introducing a new energy-based model trained for denoising with a covariance-based regularization term that enforces consistency across different measurement conditions. The trained model can compute normalized posterior densities for diverse linear inverse problems, without additional retraining or fine tuning. In addition to preserving the sampling capabilities of diffusion models, this enables previously unavailable capabilities: energy-guided adaptive sampling that adjusts schedules on-the-fly, unbiased Metropolis-Hastings correction steps, and blind estimation of the degradation operator via Bayes rule. We validate the method on multiple datasets (ImageNet, CelebA, AFHQ) and tasks (inpainting, deblurring), demonstrating competitive or superior performance to established baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2605_15487
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning Normalized Energy Models for Linear Inverse Problems
Zilberstein, Nicolas
Segarra, Santiago
Simoncelli, Eero
Guth, Florentin
Machine Learning
Computer Vision and Pattern Recognition
Image and Video Processing
Generative diffusion models can provide powerful prior probability models for inverse problems in imaging, but existing implementations suffer from two key limitations: $(i)$ the prior density is represented implicitly, and $(ii)$ they rely on likelihood approximations that introduce sampling biases. We address these challenges by introducing a new energy-based model trained for denoising with a covariance-based regularization term that enforces consistency across different measurement conditions. The trained model can compute normalized posterior densities for diverse linear inverse problems, without additional retraining or fine tuning. In addition to preserving the sampling capabilities of diffusion models, this enables previously unavailable capabilities: energy-guided adaptive sampling that adjusts schedules on-the-fly, unbiased Metropolis-Hastings correction steps, and blind estimation of the degradation operator via Bayes rule. We validate the method on multiple datasets (ImageNet, CelebA, AFHQ) and tasks (inpainting, deblurring), demonstrating competitive or superior performance to established baselines.
title Learning Normalized Energy Models for Linear Inverse Problems
topic Machine Learning
Computer Vision and Pattern Recognition
Image and Video Processing
url https://arxiv.org/abs/2605.15487