Quadratic Neural Networks for Solving Inverse Problems

被引：0

作者：

Frischauf, Leon ^{[1
]}

Scherzer, Otmar ^{[1
,2
,3
]}

Shi, Cong ^{[1
,2
,4
]}

机构：

[1] Univ Vienna, Fac Math, Vienna, Austria

[2] Johann Radon Inst Computat & Appl Math RICAM, Linz, Austria

[3] Christian Doppler Lab Math Modeling & Simulat Next, Vienna, Austria

[4] Univ Vienna, Fac Math, Oskar Morgenstern Pl 1, A-1090 Vienna, Austria

来源：

NUMERICAL FUNCTIONAL ANALYSIS AND OPTIMIZATION | 2024年 / 45卷 / 02期

基金：

奥地利科学基金会;

关键词：

Generalized neural network functions; inverse problems; MULTILAYER FEEDFORWARD NETWORKS; UNIVERSAL APPROXIMATION;

D O I：

10.1080/01630563.2024.2316580

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

In this paper we investigate the solution of inverse problems with neural network ansatz functions with generalized decision functions. The relevant observation for this work is that such functions can approximate typical test cases, such as the Shepp-Logan phantom, better, than standard neural networks. Moreover, we show that the convergence analysis of numerical methods for solving inverse problems with shallow generalized neural network functions leads to more intuitive convergence conditions, than for deep affine linear neural networks.

引用

页码：112 / 135

页数：24

共 50 条

[1] NETT: solving inverse problems with deep neural networks
Li, Housen
Schwab, Johannes
Antholzer, Stephan
Haltmeier, Markus
[J]. INVERSE PROBLEMS, 2020, 36 (06)
[2] On the recurrent neural networks for solving general quadratic programming problems
Mladenov, V
[J]. NEUREL 2004: SEVENTH SEMINAR ON NEURAL NETWORK APPLICATIONS IN ELECTRICAL ENGINEERING, PROCEEDINGS, 2004, : 5 - 9
[3] Solving Inverse Problems With Deep Neural Networks - Robustness Included?
Genzel, Martin
Macdonald, Jan
Marz, Maximilian
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2023, 45 (01) : 1119 - 1134
[4] Solving inverse problems using conditional invertible neural networks
Padmanabha, Govinda Anantha
Zabaras, Nicholas
[J]. JOURNAL OF COMPUTATIONAL PHYSICS, 2021, 433
[5] Solving inverse problems by Bayesian iterative inversion of Neural Networks
Hwang, JN
[J]. THEORETICAL ASPECTS OF NEURAL COMPUTATION: A MULTIDISCIPLINARY PERSPECTIVE, 1998, : 103 - 117
[6] Solving groundwater inverse problems using artificial neural networks
Rizzo, DM
Dougherty, DE
[J]. COMPUTATIONAL METHODS IN WATER RESOURCES XI, VOL 1: COMPUTATIONAL METHODS IN SUBSURFACE FLOW AND TRANSPORT PROBLEMS, 1996, : 313 - 319
[7] ARTIFICIAL NEURAL NETWORKS FOR SOLVING OF OPTICS INVERSE AND DIRECT PROBLEMS
Abrukov, Victor S.
[J]. DETC 2008: PROCEEDINGS OF THE ASME INTERNATIONAL DESIGN ENGINEERING TECHNICAL CONFERENCES AND COMPUTERS AND INFORMATIONAL IN ENGINEERING CONFERENCE, VOL 3, PTS A AND B: 28TH COMPUTERS AND INFORMATION IN ENGINEERING CONFERENCE, 2009, : 1043 - 1047
[8] DATA-CONSISTENT NEURAL NETWORKS FOR SOLVING NONLINEAR INVERSE PROBLEMS
Boink, Yoeri E.
Haltmeier, Markus
Holman, Sean
Schwab, Johannes
[J]. INVERSE PROBLEMS AND IMAGING, 2023, 17 (01) : 203 - 229
[9] Finite element interpolated neural networks for solving forward and inverse problems
Badia, Santiago
Li, Wei
Martin, Alberto F.
[J]. COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING, 2024, 418
[10] Training RBF neural networks for solving nonlinear and inverse boundary value problems
Jankowska, Malgorzata A.
Karageorghis, Andreas
Chen, C. S.
[J]. COMPUTERS & MATHEMATICS WITH APPLICATIONS, 2024, 165 : 205 - 216

← 1 2 3 4 5 →