Stability Bounds for the Unfolded Forward-Backward Algorithm

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Chouzenoux, Emilie, Della Valle, Cecile, Pesquet, Jean-Christophe
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912619320836096
author Chouzenoux, Emilie
Della Valle, Cecile
Pesquet, Jean-Christophe
author_facet Chouzenoux, Emilie
Della Valle, Cecile
Pesquet, Jean-Christophe
contents We consider a neural network architecture designed to solve inverse problems where the degradation operator is linear and known. This architecture is constructed by unrolling a forward-backward algorithm derived from the minimization of an objective function that combines a data-fidelity term, a Tikhonov-type regularization term, and a potentially nonsmooth convex penalty. The robustness of this inversion method to input perturbations is analyzed theoretically. Ensuring robustness complies with the principles of inverse problem theory, as it ensures both the continuity of the inversion method and the resilience to small noise - a critical property given the known vulnerability of deep neural networks to adversarial perturbations. A key novelty of our work lies in examining the robustness of the proposed network to perturbations in its bias, which represents the observed data in the inverse problem. Additionally, we provide numerical illustrations of the analytical Lipschitz bounds derived in our analysis.
format Preprint
id arxiv_https___arxiv_org_abs_2412_17888
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Stability Bounds for the Unfolded Forward-Backward Algorithm
Chouzenoux, Emilie
Della Valle, Cecile
Pesquet, Jean-Christophe
Optimization and Control
Artificial Intelligence
We consider a neural network architecture designed to solve inverse problems where the degradation operator is linear and known. This architecture is constructed by unrolling a forward-backward algorithm derived from the minimization of an objective function that combines a data-fidelity term, a Tikhonov-type regularization term, and a potentially nonsmooth convex penalty. The robustness of this inversion method to input perturbations is analyzed theoretically. Ensuring robustness complies with the principles of inverse problem theory, as it ensures both the continuity of the inversion method and the resilience to small noise - a critical property given the known vulnerability of deep neural networks to adversarial perturbations. A key novelty of our work lies in examining the robustness of the proposed network to perturbations in its bias, which represents the observed data in the inverse problem. Additionally, we provide numerical illustrations of the analytical Lipschitz bounds derived in our analysis.
title Stability Bounds for the Unfolded Forward-Backward Algorithm
topic Optimization and Control
Artificial Intelligence
url https://arxiv.org/abs/2412.17888