The residual method for regularizing ill-posed problems
From MaRDI portal
\(L_p\) spaceBanach spacecompressed sensingconvergenceill-posed problemsimage processingresidual methodsignal processingsparsitystabilityTikhonov regularization
Nonlinear ill-posed problems (47J06) Numerical solutions to equations with nonlinear operators (65J15) Numerical solutions of ill-posed problems in abstract spaces; regularization (65J20) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)
Abstract: Although the emph{residual method}, or emph{constrained regularization}, is frequently used in applications, a detailed study of its properties is still missing. This sharply contrasts the progress of the theory of Tikhonov regularization, where a series of new results for regularization in Banach spaces has been published in the recent years. The present paper intends to bridge the gap between the existing theories as far as possible. We develop a stability and convergence theory for the residual method in general topological spaces. In addition, we prove convergence rates in terms of (generalized) Bregman distances, which can also be applied to non-convex regularization functionals. We provide three examples that show the applicability of our theory. The first example is the regularized solution of linear operator equations on -spaces, where we show that the results of Tikhonov regularization generalize unchanged to the residual method. As a second example, we consider the problem of density estimation from a finite number of sampling points, using the Wasserstein distance as a fidelity term and an entropy measure as regularization term. It is shown that the densities obtained in this way depend continuously on the location of the sampled points and that the underlying density can be recovered as the number of sampling points tends to infinity. Finally, we apply our theory to compressed sensing. Here, we show the well-posedness of the method and derive convergence rates both for convex and non-convex regularization under rather weak conditions.
In signal or image processing (particularly, in compressed sensing) the main considered tool is the residual method. A deeper mathematical frame within Banach or topological vectorial spaces is still missing in contrast to the equivalent (at some point) Tikhonov regularization. Drawing a parallel between these two methods, we can imagine the missing mathematical gap within the residual method. The paper aims to fill that lack ``as much as possible. Specifically, the existence (well-posedness), stability and convergence (rates) are discussed. Three main examples are given. First, one deals with regularization on \(L_p\)-spaces and shows the equivalence between Tikhonov and residual methods. Secondly, one deals with regularization of probability measures. The last example regards the compressed sensing.
Recommendations
- On the connection between the generalized discrepancy method and the generalized discrepancy principle for non-linear ill-posed problems
- scientific article; zbMATH DE number 3926111
- A convergence criterion for approximations in the residual method for linear ill-posed problems
- scientific article; zbMATH DE number 484354
- Regularization Methods for Ill-Posed Problems
Cites work
- A class of Wasserstein metrics for probability distributions
- A convergence rates result for Tikhonov regularization in Banach spaces with non-smooth operators
- A New Approach to Source Conditions in Regularization with General Residual Term
- A variational formulation for frame-based inverse problems
- An extension of the variational inequality approach for obtaining convergence rates in regularization of nonlinear ill-posed problems
- An Introduction to Banach Space Theory
- An iterative thresholding algorithm for linear inverse problems with a sparsity constraint
- Analysis of bounded variation penalty methods for ill-posed problems
- Approximate source conditions for nonlinear ill-posed problems—chances and limitations
- Characteristic inequalities of uniformly convex and uniformly smooth Banach spaces
- Convergence criterion of the residual method
- Convergence rates and source conditions for Tikhonov regularization with sparsity constraints
- Convergence rates for Tikhonov regularisation of non-linear ill-posed problems
- Convergence rates for Tikhonov regularization based on range inclusions
- Convergence rates of convex variational regularization
- Error estimates for non-quadratic regularization and the relation to enhancement
- For most large underdetermined systems of equations, the minimal 𝓁1‐norm near‐solution approximates the sparsest near‐solution
- For most large underdetermined systems of linear equations the minimal 𝓁1‐norm solution is also the sparsest solution
- Generalized Bregman distances and convergence rates for non-convex regularization methods
- scientific article; zbMATH DE number 3112450 (Why is no real title available?)
- scientific article; zbMATH DE number 3868512 (Why is no real title available?)
- scientific article; zbMATH DE number 5289889 (Why is no real title available?)
- scientific article; zbMATH DE number 3770434 (Why is no real title available?)
- scientific article; zbMATH DE number 3770864 (Why is no real title available?)
- scientific article; zbMATH DE number 48239 (Why is no real title available?)
- scientific article; zbMATH DE number 3504682 (Why is no real title available?)
- scientific article; zbMATH DE number 3551792 (Why is no real title available?)
- scientific article; zbMATH DE number 3626044 (Why is no real title available?)
- scientific article; zbMATH DE number 1197455 (Why is no real title available?)
- scientific article; zbMATH DE number 1017537 (Why is no real title available?)
- scientific article; zbMATH DE number 1865939 (Why is no real title available?)
- scientific article; zbMATH DE number 872525 (Why is no real title available?)
- scientific article; zbMATH DE number 936298 (Why is no real title available?)
- scientific article; zbMATH DE number 3255438 (Why is no real title available?)
- scientific article; zbMATH DE number 3307878 (Why is no real title available?)
- Just relax: convex programming methods for identifying sparse signals in noise
- Minimization of Tikhonov functionals in Banach spaces
- Modern methods in the calculus of variations. L^p spaces
- Near-Optimal Signal Recovery From Random Projections: Universal Encoding Strategies?
- Necessary and sufficient conditions for linear convergence of ^1-regularization
- Non-convex sparse regularisation
- On Tikhonov regularization with non-convex sparsity constraints
- Optimal a Posteriori Parameter Choice for Tikhonov Regularization for Solving Nonlinear Ill-Posed Problems
- Recovery of Exact Sparse Representations in the Presence of Bounded Noise
- Regularization of ill-posed problems in Banach spaces: convergence rates
- Regularization with non-convex separable constraints
- Restricted Isometry Constants Where $\ell ^{p}$ Sparse Recovery Can Fail for $0≪ p \leq 1$
- Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
- Some tendencies in the Tikhonov regularization of ill-posed problems
- Sparse and redundant representations. From theory to applications in signal and image processing.
- Sparse regularization with l q penalty term
- Sparsest solutions of underdetermined linear systems via \( \ell _q\)-minimization for \(0<q\leqslant 1\)
- Stable recovery of sparse overcomplete representations in the presence of noise
- Stable signal recovery from incomplete and inaccurate measurements
- Strong convergence results related to strict convexity
- The restricted isometry property and its implications for compressed sensing
- Variational methods in imaging
- Well posedness and convergence of some regularisation methods for non-linear ill posed problems
- Well-posedness and convergence rates for sparse regularization with sublinear \(l^q\) penalty term
Cited in
(21)- Convergence and adaptive discretization of the IRGNM Tikhonov and the IRGNM Ivanov method under a tangential cone condition in Banach space
- Existence of variational source conditions for nonlinear inverse problems in Banach spaces
- Infinite-dimensional inverse problems with finite measurements
- Concentration of \(S\)-largest mutilated vectors with \(\ell_p\)-quasinorm for \(0<p\leq 1\) and its applications
- Infinite dimensional compressed sensing from anisotropic measurements and applications to inverse problems in PDE
- The elementary residual method
- Regularization Methods for Ill-Posed Problems
- A TGV-based framework for variational image decompression, zooming, and reconstruction. I: Analytics
- Diffusion tensor imaging with deterministic error bounds
- A New Approach to Source Conditions in Regularization with General Residual Term
- Joint additive Kullback–Leibler residual minimization and regularization for linear inverse problems
- scientific article; zbMATH DE number 3905262 (Why is no real title available?)
- scientific article; zbMATH DE number 3956335 (Why is no real title available?)
- On convergence and convergence rates for Ivanov and Morozov regularization and application to some parameter identification problems in elliptic PDEs
- Image reconstruction with imperfect forward models and applications in deblurring
- Regularization graphs -- a unified framework for variational regularization of inverse problems
- NETT: solving inverse problems with deep neural networks
- Variational regularisation for inverse problems with imperfect forward operators and general noise models
- Solving an inverse source problem by deep neural network method with convergence and error analysis
- Data-Driven Morozov Regularization of Inverse Problems
- Nested Bregman iterations for decomposition problems
This page was built for publication: The residual method for regularizing ill-posed problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q425458)
