Deep Equilibrium Models for Poisson Imaging Inverse Problems via Mirror Descent
SIAM Journal on Imaging Sciences · 30.06.2026 · 19 (2) · 1077-1109
DOI: 10.1137/25m1779048 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Imaging Sciences, online erschienen am 12.05.2026. Verfasst von Christian Daniele und 3 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 53 Literaturangaben. Gefördert von: Agence Nationale de la Recherche, H2020 European Research Council, European Commission.
Abstract (Originalsprache)
Deep equilibrium models (DEQs) are implicit neural networks with fixed points that have recently gained attention for learning image regularization functionals, particularly in settings involving Gaussian fidelities, where assumptions on the forward operator ensure contractiveness of standard (proximal) gradient descent operators. In this work, we extend the application of DEQs to Poisson inverse problems, where the data fidelity term is more appropriately modeled by the Kullback–Leibler divergence. To this end, we introduce a novel DEQ formulation based on mirror descent defined in terms of a tailored non-Euclidean geometry that naturally adapts with the structure of the data term. This enables the learning of neural regularizers within a principled training framework. We derive sufficient conditions and establish refined convergence results based on the Kurdyka–Łojasiewicz framework for functions with nonclosed domains to guarantee the convergence of the learned reconstruction scheme and propose computational strategies that enable both efficient training and parameter-free inference. Numerical experiments show that our method outperforms traditional model-based approaches, and it is comparable to the performance of Bregman plug-and-play methods, while mitigating their typical drawbacks, such as time-consuming tuning of hyperparameters. The code is publicly available at https://github.com/christiandaniele/DEQ-MD .
Förderung
- Agence Nationale de la Recherche
- H2020 European Research Council
- European Commission
- Air Force Office of Scientific Research
- Ministry of Education, University and Research
- PNRR
- Ministero dell’Istruzione, dell’Università e della Ricerca
- Gruppo Nazionale per l’Analisi Matematica, la Probabilità e le loro Applicazioni
Zitieren
APA
Daniele, C., Villa, S., Vaiter, S., & Calatroni, L. (2026). Deep Equilibrium Models for Poisson Imaging Inverse Problems via Mirror Descent. SIAM Journal on Imaging Sciences, 19(2), 1077-1109. https://doi.org/10.1137/25m1779048
Vancouver
Daniele C, Villa S, Vaiter S, Calatroni L. Deep Equilibrium Models for Poisson Imaging Inverse Problems via Mirror Descent. SIAM Journal on Imaging Sciences. 2026;19(2):1077-1109. doi:10.1137/25m1779048
BibTeX
@article{Daniele2026,
author = {Daniele, Christian and Villa, Silvia and Vaiter, Samuel and Calatroni, Luca},
title = {Deep Equilibrium Models for Poisson Imaging Inverse Problems via Mirror Descent},
journal = {SIAM Journal on Imaging Sciences},
year = {2026},
volume = {19},
number = {2},
pages = {1077--1109},
doi = {10.1137/25m1779048}
}
EUVIMED