A Framework for Regularization via Operator Approximation

  title={A Framework for Regularization via Operator Approximation},
  author={Julianne Chung and Misha Elena Kilmer and Dianne P. O’Leary},
  journal={SIAM J. Sci. Comput.},
Regularization approaches based on spectral filtering can be highly effective in solving ill-posed inverse problems. These methods, however, require computing the singular value decomposition (SVD) and choosing appropriate regularization parameters. These tasks can be prohibitively expensive for large-scale problems. In this paper, we present a framework that uses operator approximations to efficiently obtain good regularization parameters without an SVD of the original operator. Instead, we… 

Optimal Regularization Parameters for General-Form Tikhonov Regularization

In this work we consider the problem of finding optimal regularization parameters for general-form Tikhonov regularization using training data. We formulate the general-form Tikhonov solution as a

Learning regularization parameters for general-form Tikhonov

An efficient learning approach for computing regularization parameters that can be used for a large set of problems, and a new class of regularizing filters that requires less data than previously proposed optimal error filters, avoids the generalized SVD, and allows flexibility and novelty in the choice of regularization matrices.

Near-Optimal Spectral Filtering and Error Estimation for Solving Ill-Posed Problems

This work considers regularization methods for numerical solution of linear ill-posed problems, in particular image deblurring, when the singular value decomposition (SVD) of the operator is available, and defines the Picard parameter, the index beyond which the data are dominated by noise.

Computational methods for large-scale inverse problems: a survey on hybrid projection methods

This paper surveys an important class of methods that combine iterative projection methods and variational regularization methods for large-scale inverse problems, providing researchers with a powerful computational framework for solving very large inverse problems.

Adaptive Regularization Parameter Choice Rules for Large-Scale Problems

This paper derives a new class of adaptive regularization parameter choice strategies that can be effectively and efficiently applied when regularizing large-scale linear inverse problems by

Krylov methods for inverse problems: Surveying classical, and introducing new, algorithmic approaches

Large‐scale linear systems coming from suitable discretizations of linear inverse problems are challenging to solve. Indeed, since they are inherently ill‐posed, appropriate regularization should be

Learning multiple regularization parameters for generalized Tikhonov regularization using multiple data sets without true data

This paper considers the unbiased risk estimator, generalized cross validation, and the discrepancy principle as the means of selecting regularization parameters and demonstrates that these modified methods are effective and efficient, and perform comparably to methods based on true data for learning the relevant parameters.

An inner–outer iterative method for edge preservation in image restoration and reconstruction

The value of the approach on applications in X-ray CT image reconstruction and in image deblurring is demonstrated, showing that it can be computationally much more attractive than other well-known strategies for edge preservation, while providing solutions of greater or equal quality.

Regularization properties of LSQR for linear discrete ill-posed problems in the multiple singular value case and best, near best and general low rank approximations

For the large-scale linear discrete ill-posed problem min‖Ax − b‖ or Ax = b with b contaminated by white noise, the Golub–Kahan bidiagonalization based LSQR method and its mathematically equivalent

Regularization by inexact Krylov methods with applications to blind deblurring

Two new inexact Krylov methods are derived that can be efficiently applied to unregularized or Tikhonov-regularized least squares problems, and their theoretical properties are studied, including links with their exact counterparts and strategies to monitor the amount of inexactness.



Choosing Regularization Parameters in Iterative Methods for Ill-Posed Problems

This work presents a common framework for efficient algorithms that regularize after this second projection rather than before it, and shows that determining regularization parameters based on the final projected problem rather than on the original discretization has firmer justification and often involves less computational expense.

Windowed Spectral Regularization of Inverse Problems

This work shows how to use standard parameter-choice methods, such as the discrepancy principle and generalized cross-validation, in a windowed regularization framework, and demonstrates the effectiveness of the algorithms on deblurring images and on the backward heat equation.

Near-Optimal Parameters for Tikhonov and Other Regularization Methods

This work proposes choosing the parameter, without a priori information, by approximately minimizing the distance between the true solution to the discrete problem and the family of regularized solutions and proves convergence of the regularization parameter to zero as the standard deviation of the noise goes to zero.

REGULARIZATION TOOLS: A Matlab package for analysis and solution of discrete ill-posed problems

  • P. Hansen
  • Mathematics, Computer Science
    Numerical Algorithms
  • 2005
The package REGULARIZATION TOOLS consists of 54 Matlab routines for analysis and solution of discrete ill-posed problems, i.e., systems of linear equations whose coefficient matrix has the properties


A weighted-GCV method for Lanczos-hybrid regularization.

A weighted generalized cross validation (WGCV) method is described that the semi-convergence behavior of the Lanczos method can be overcome, making the solution less sensitive to the number of iterations.

A Bidiagonalization-Regularization Procedure for Large Scale Discretizations of Ill-Posed Problems

In this paper, ill-posed problems which discretize to linear least squares problems with matrices K of high dimensions are considered and an algorithm proposed which uses K only as an operator and does not need to explicitly store or modify it.

A New Alternating Minimization Algorithm for Total Variation Image Reconstruction

An alternating minimization algorithm for recovering images from blurry and noisy observations with total variation (TV) regularization from a new half-quadratic model applicable to not only the anisotropic but also the isotropic forms of TV discretizations is proposed.

Tikhonov Regularization of Large Linear Problems

Many numerical methods for the solution of linear ill-posed problems apply Tikhonov regularization. This paper presents a new numerical method, based on Lanczos bidiagonalization and Gauss