A Data-Driven Iteratively Regularized Landweber Iteration

Abstract We derive and analyze a new variant of the iteratively regularized Landweber iteration, for solving linear and nonlinear ill-posed inverse problems. The method takes into account training data, which are used to estimate the interior of a black box, which is used to define the iteration process. We prove convergence and stability for the scheme in infinite dimensional Hilbert spaces. These theoretical results are complemented by some numerical experiments for solving linear inverse problems for the Radon transform and a nonlinear inverse problem for Schlieren tomography.

[1]  A. Bakushinskii.  The problem of the convergence of the iteratively regularized Gauss-Newton method , 1992 .

[2]  Otmar Scherzer,et al.  KACZMARZ METHODS FOR REGULARIZING NONLINEAR ILL-POSED EQUATIONS I: CONVERGENCE ANALYSIS , 2007 .

[3]  M. Haltmeier,et al.  Regularization of systems of nonlinear ill-posed equations: II. Applications , 2020, ArXiv.

[4]  Otmar Scherzer,et al.  Variational Methods in Imaging , 2008, Applied mathematical sciences.

[5]  E. Bolinder.  The Fourier integral and its applications , 1963 .

[6]  Peter Kuchment,et al.  The Radon Transform and Medical Imaging , 2014, CBMS-NSF regional conference series in applied mathematics.

[7]  Eric Todd Quinto,et al.  The Radon Transform, Inverse Problems, and Tomography , 2006 .

[8]  Otmar Scherzer,et al.  KACZMARZ METHODS FOR REGULARIZING NONLINEAR ILL-POSED EQUATIONS II: APPLICATIONS , 2007 .

[9]  Stephan Antholzer,et al.  Deep learning for photoacoustic tomography from sparse data , 2017, Inverse problems in science and engineering.

[11]  Thomas G. Dietterich.  Adaptive computation and machine learning , 1998 .

[12]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[13]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[14]  Barbara Kaltenbacher,et al.  Iterative Regularization Methods for Nonlinear Ill-Posed Problems , 2008, Radon Series on Computational and Applied Mathematics.

[15]  Otmar Scherzer,et al.  A Modified Landweber Iteration for Solving Parameter Estimation Problems , 1998 .

[16]  Markus Haltmeier,et al.  Operator Learning Approach for the Limited View Problem in Photoacoustic Tomography , 2019, Comput. Methods Appl. Math..

[17]  Jonas Adler,et al.  Solving ill-posed inverse problems using iterative deep neural networks , 2017, ArXiv.

[18]  Wojciech Samek,et al.  Learning the invisible: a hybrid deep learning-shearlet framework for limited angle computed tomography , 2018, Inverse Problems.

[19]  Tosio Kato.  Perturbation theory for linear operators , 1966 .

[20]  David Lindley,et al.  CBMS-NSF REGIONAL CONFERENCE SERIES IN APPLIED MATHEMATICS , 2010 .

[21]  A. Papoulis,et al.  The Fourier Integral and Its Applications , 1963 .

[22]  J. Greenleaf,et al.  Tomographic Schlieren imaging for measurement of beam pressure and intensity , 1994, 1994 Proceedings of IEEE Ultrasonics Symposium.

[23]  Barbara Kaltenbacher,et al.  Regularization Methods in Banach Spaces , 2012, Radon Series on Computational and Applied Mathematics.