On Learned Operator Correction in Inverse Problems

DOI10.1137/20M1338460zbMath1474.65182arXiv2005.07069OpenAlexW3122711257MaRDI QIDQ5860277

Sebastian Lunz, Andreas Hauptmann, Tanja Tarvainen, Carola-Bibiane Schönlieb, Simon R. Arridge

Publication date: 19 November 2021

Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2005.07069

zbMATH Keywords

inverse problemsvariational methodsdeep learningphotoacoustic tomographymodel correctionoperator learning

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Ill-posedness and regularization problems in numerical linear algebra (65F22) Numerical optimization and variational techniques (65K10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Linear operators and ill-posed problems, regularization (47A52)

Related Items

Operator Shifting for General Noisy Matrix Systems, On global normal linear approximations for nonlinear Bayesian inverse problems, Learning and correcting non-Gaussian model errors, Deep learning for inverse problems. Abstracts from the workshop held March 7--13, 2021 (hybrid meeting), Divide-and-conquer DNN approach for the inverse point source problem using a few single frequency measurements, In focus -- hybrid deep learning approaches to the HDC2021 challenge, Operator shifting for noisy elliptic systems, Sequential Model Correction for Nonlinear Inverse Problems, Utilizing Variational Autoencoders in the Bayesian Inverse Problem of Photoacoustic Tomography, Proximal nested sampling for high-dimensional Bayesian model selection

Uses Software

• k-Wave
• U-Net
• DeepAdverserialRegulariser
• NETT

Cites Work

• Statistical and computational inverse problems.
• Less is often more: applied inverse problems using \(hp\)-forward models
• Statistical inverse problems: discretization, model reduction and inverse crimes
• Bayesian Calibration of Computer Models
• Approximation errors and model reduction with an application in optical diffusion tomography
• Mathematics of thermoacoustic tomography
• Model reduction methods based on Krylov subspaces
• Solving ill-posed inverse problems using iterative deep neural networks
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Convergence rates and structure of solutions of inverse problems with imperfect forward models
• Deep null space learning for inverse problems: convergence analysis and rates
• Image Reconstruction with Imperfect Forward Models and Applications in Deblurring
• Deep learning for photoacoustic tomography from sparse data
• Reduced order models for spectral domain inversion: embedding into the continuous problem and generation of internal data
• NETT: solving inverse problems with deep neural networks
• Parameter identification in a structured population model
• Solving inverse problems using data-driven models
• Identification of Chemotaxis Models with Volume-Filling
• Learning the invisible: a hybrid deep learning-shearlet framework for limited angle computed tomography