Hybrid and Iteratively Reweighted Regularization by Unbiased Predictive Risk and Weighted GCV for Projected Systems

DOI10.1137/15M1037925zbMath1360.65115arXiv1509.00096MaRDI QIDQ2968588

Saeed Vatankhah, Vahid E. Ardestani, Rosemary A. Renaut

Publication date: 20 March 2017

Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1509.00096

zbMATH Keywords

numerical examplesdiscrepancy principleimage reconstructiongeneralized cross validationimage restorationGolub-Kahan bidiagonalizationregularization parameter estimationlarge-scale inverse problemsiteratively reweighted schemesunbiased predictive risk estimator

Mathematics Subject Classification ID

Ill-posedness and regularization problems in numerical linear algebra (65F22) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18) Iterative numerical methods for linear systems (65F10)

Related Items (16)

Learning spectral windowing parameters for regularization using unbiased predictive risk and generalized cross validation techniques for multiple data sets ⋮ A computational framework for edge-preserving regularization in dynamic inverse problems ⋮ A relaxed iterated Tikhonov regularization for linear ill-posed inverse problems ⋮ A joint bidiagonalization based iterative algorithm for large scale general-form Tikhonov regularization ⋮ Krylov methods for inverse problems: Surveying classical, and introducing new, algorithmic approaches ⋮ Double precision is not necessary for LSQR for solving discrete linear ill-posed problems ⋮ 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 ⋮ Efficient generalized Golub–Kahan based methods for dynamic inverse problems ⋮ Generalized Hybrid Iterative Methods for Large-Scale Bayesian Inverse Problems ⋮ An inner–outer iterative method for edge preservation in image restoration and reconstruction ^* ⋮ Krylov Methods for Low-Rank Regularization ⋮ Unbiased predictive risk estimation of the Tikhonov regularization parameter: convergence with increasing rank approximations of the singular value decomposition ⋮ Sampled Tikhonov regularization for large linear inverse problems ⋮ Approximation accuracy of the Krylov subspaces for linear discrete ill-posed problems ⋮ Flexible Krylov Methods for $\ell_p$ Regularization ⋮ Hybrid projection methods for large-scale inverse problems with mixed Gaussian priors

Uses Software

• Matlab
• RestoreTools
• LSMR

Cites Work

• Unnamed Item
• Unnamed Item
• Tikhonov regularization based on generalized Krylov subspace methods
• A weighted-GCV method for Lanczos-hybrid regularization
• Noise propagation in regularizing iterations for image deblurring
• Regularization methods for large-scale problems
• Some results on the regularization of LSQR for large-scale discrete ill-posed problems
• Iterative methods for image deblurring: A Matlab object-oriented approach
• Regularization parameter estimation for large-scale Tikhonov regularization using a priori information
• The regularizing effect of the Golub-Kahan iterative bidiagonalization and revealing the noise level in the data
• Choosing Regularization Parameters in Iterative Methods for Ill-Posed Problems
• Sparse Tomography
• LSMR: An Iterative Algorithm for Sparse Least-Squares Problems
• Windowed Spectral Regularization of Inverse Problems
• Regularization parameter estimation for underdetermined problems by the χ ² principle with application to 2D focusing gravity inversion
• Deblurring Images
• Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
• Towards a Generalized Singular Value Decomposition
• LSQR: An Algorithm for Sparse Linear Equations and Sparse Least Squares
• Computational Methods for Inverse Problems
• A Framework for Regularization via Operator Approximation
• ForWaRD: Fourier-Wavelet Regularized Deconvolution for Ill-Conditioned Systems

This page was built for publication: Hybrid and Iteratively Reweighted Regularization by Unbiased Predictive Risk and Weighted GCV for Projected Systems