Guardado en:
Detalles Bibliográficos
Autores principales: Buskulic, Nathan, Fadili, Jalal, Quéau, Yvain
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:https://arxiv.org/abs/2309.12128
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909138169102336
author Buskulic, Nathan
Fadili, Jalal
Quéau, Yvain
author_facet Buskulic, Nathan
Fadili, Jalal
Quéau, Yvain
contents Neural networks have become a prominent approach to solve inverse problems in recent years. While a plethora of such methods was developed to solve inverse problems empirically, we are still lacking clear theoretical guarantees for these methods. On the other hand, many works proved convergence to optimal solutions of neural networks in a more general setting using overparametrization as a way to control the Neural Tangent Kernel. In this work we investigate how to bridge these two worlds and we provide deterministic convergence and recovery guarantees for the class of unsupervised feedforward multilayer neural networks trained to solve inverse problems. We also derive overparametrization bounds under which a two-layers Deep Inverse Prior network with smooth activation function will benefit from our guarantees.
format Preprint
id arxiv_https___arxiv_org_abs_2309_12128
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Convergence and Recovery Guarantees of Unsupervised Neural Networks for Inverse Problems
Buskulic, Nathan
Fadili, Jalal
Quéau, Yvain
Machine Learning
Neural networks have become a prominent approach to solve inverse problems in recent years. While a plethora of such methods was developed to solve inverse problems empirically, we are still lacking clear theoretical guarantees for these methods. On the other hand, many works proved convergence to optimal solutions of neural networks in a more general setting using overparametrization as a way to control the Neural Tangent Kernel. In this work we investigate how to bridge these two worlds and we provide deterministic convergence and recovery guarantees for the class of unsupervised feedforward multilayer neural networks trained to solve inverse problems. We also derive overparametrization bounds under which a two-layers Deep Inverse Prior network with smooth activation function will benefit from our guarantees.
title Convergence and Recovery Guarantees of Unsupervised Neural Networks for Inverse Problems
topic Machine Learning
url https://arxiv.org/abs/2309.12128