Let data talk: data-regularized operator learning theory for inverse problems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Ke, Wang, Chunmei, Yang, Haizhao
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911806266540032
author Chen, Ke
Wang, Chunmei
Yang, Haizhao
author_facet Chen, Ke
Wang, Chunmei
Yang, Haizhao
contents Regularization plays a pivotal role in integrating prior information into inverse problems. While many deep learning methods have been proposed to solve inverse problems, determining where to apply regularization remains a crucial consideration. Typical methods regularize neural networks via architecture, wherein neural network functions parametrize the parameter of interest or the regularization term. We introduce a novel approach, denoted as the "data-regularized operator learning" (DaROL) method, designed to address PDE inverse problems. The DaROL method trains a neural network on data, regularized through common techniques such as Tikhonov variational methods and Bayesian inference. The DaROL method offers flexibility across different frameworks, faster inverse problem-solving, and a simpler structure that separates regularization and neural network training. We demonstrate that training a neural network on the regularized data is equivalent to supervised learning for a regularized inverse map. Furthermore, we provide sufficient conditions for the smoothness of such a regularized inverse map and estimate the learning error in terms of neural network size and the number of training samples.
format Preprint
id arxiv_https___arxiv_org_abs_2310_09854
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Let data talk: data-regularized operator learning theory for inverse problems
Chen, Ke
Wang, Chunmei
Yang, Haizhao
Numerical Analysis
65N21
G.1.8
Regularization plays a pivotal role in integrating prior information into inverse problems. While many deep learning methods have been proposed to solve inverse problems, determining where to apply regularization remains a crucial consideration. Typical methods regularize neural networks via architecture, wherein neural network functions parametrize the parameter of interest or the regularization term. We introduce a novel approach, denoted as the "data-regularized operator learning" (DaROL) method, designed to address PDE inverse problems. The DaROL method trains a neural network on data, regularized through common techniques such as Tikhonov variational methods and Bayesian inference. The DaROL method offers flexibility across different frameworks, faster inverse problem-solving, and a simpler structure that separates regularization and neural network training. We demonstrate that training a neural network on the regularized data is equivalent to supervised learning for a regularized inverse map. Furthermore, we provide sufficient conditions for the smoothness of such a regularized inverse map and estimate the learning error in terms of neural network size and the number of training samples.
title Let data talk: data-regularized operator learning theory for inverse problems
topic Numerical Analysis
65N21
G.1.8
url https://arxiv.org/abs/2310.09854