A deep first-order system least squares method for the obstacle problem

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Acosta, Gabriel, Belén, Eugenia, Bersetche, Francisco M., Borthagaray, Juan Pablo
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915465917366272
author Acosta, Gabriel
Belén, Eugenia
Bersetche, Francisco M.
Borthagaray, Juan Pablo
author_facet Acosta, Gabriel
Belén, Eugenia
Bersetche, Francisco M.
Borthagaray, Juan Pablo
contents We propose a deep learning approach to the obstacle problem inspired by the first-order system least-squares (FOSLS) framework. This method reformulates the problem as a convex minimization task; by simultaneously approximating the solution, gradient, and Lagrange multiplier, our approach provides a flexible, mesh-free alternative that scales efficiently to high-dimensional settings. Key theoretical contributions include the coercivity and local Lipschitz continuity of the proposed least-squares functional, along with convergence guarantees via $Γ$-convergence theory under mild regularity assumptions. Numerical experiments in dimensions up to 20 demonstrate the method's robustness and scalability, even on non-Lipschitz domains.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19412
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A deep first-order system least squares method for the obstacle problem
Acosta, Gabriel
Belén, Eugenia
Bersetche, Francisco M.
Borthagaray, Juan Pablo
Numerical Analysis
We propose a deep learning approach to the obstacle problem inspired by the first-order system least-squares (FOSLS) framework. This method reformulates the problem as a convex minimization task; by simultaneously approximating the solution, gradient, and Lagrange multiplier, our approach provides a flexible, mesh-free alternative that scales efficiently to high-dimensional settings. Key theoretical contributions include the coercivity and local Lipschitz continuity of the proposed least-squares functional, along with convergence guarantees via $Γ$-convergence theory under mild regularity assumptions. Numerical experiments in dimensions up to 20 demonstrate the method's robustness and scalability, even on non-Lipschitz domains.
title A deep first-order system least squares method for the obstacle problem
topic Numerical Analysis
url https://arxiv.org/abs/2508.19412