Sampled Tikhonov regularization for large linear inverse problems
From MaRDI portal
Abstract: In this paper, we investigate iterative methods that are based on sampling of the data for computing Tikhonov-regularized solutions. We focus on very large inverse problems where access to the entire data set is not possible all at once (e.g., for problems with streaming or massive datasets). Row-access methods provide an ideal framework for solving such problems, since they only require access to "blocks" of the data at any given time. However, when using these iterative sampling methods to solve inverse problems, the main challenges include a proper choice of the regularization parameter, appropriate sampling strategies, and a convergence analysis. To address these challenges, we first describe a family of sampled iterative methods that can incorporate data as they become available (e.g., randomly sampled). We consider two sampled iterative methods, where the iterates can be characterized as solutions to a sequence of approximate Tikhonov problems. The first method requires the regularization parameter to be fixed a priori and converges asymptotically to an unregularized solution for randomly sampled data. This is undesirable for inverse problems. Thus, we focus on the second method where the main benefits are that the regularization parameter can be updated during the iterative process and the iterates converge asymptotically to a Tikhonov-regularized solution. We describe adaptive approaches to update the regularization parameter that are based on sampled residuals, and we describe a limited-memory variant for larger problems. Numerical examples, including a large-scale super-resolution imaging example, demonstrate the potential for these methods.
Recommendations
- Generalized Tikhonov regularization method for large-scale linear inverse problems
- Linear inverse problems and Tikhonov regularization
- Randomized algorithms for large-scale inverse problems with general Tikhonov regularizations
- scientific article; zbMATH DE number 1304508
- Sampled limited memory methods for massive linear inverse problems
- Tikhonov regularization of large linear problems
- Iterative solvers for Tikhonov regularization of dense inverse problems
- An alternative to Tikhonov regularization for linear sampling methods
- Large-scale Tikhonov regularization of total least squares
- Greedy Tikhonov regularization for large linear ill-posed problems
Cites work
- A data-scalable randomized misfit approach for solving large-scale PDE-constrained inverse problems
- A randomized Kaczmarz algorithm with exponential convergence
- A stochastic quasi-Newton method for large-scale optimization
- Alternating projection methods.
- An effective method for parameter estimation with PDE constraints with multiple right-hand sides
- An introduction to data analysis and uncertainty quantification for inverse problems
- Choosing regularization parameters in iterative methods for ill-posed problems
- Computational uncertainty quantification for inverse problems
- De-noising by soft-thresholding
- Deblurring Images
- Discrete inverse problems. Insight and algorithms.
- Fast approximation of matrix coherence and statistical leverage
- Faster least squares approximation
- Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
- Generalized row-action methods for tomographic imaging
- Global convergence of online limited memory BFGS
- scientific article; zbMATH DE number 1043533 (Why is no real title available?)
- scientific article; zbMATH DE number 852536 (Why is no real title available?)
- scientific article; zbMATH DE number 5060482 (Why is no real title available?)
- scientific article; zbMATH DE number 3027894 (Why is no real title available?)
- Hybrid and iteratively reweighted regularization by unbiased predictive risk and weighted GCV for projected systems
- Iterated Tikhonov regularization with a general penalty term.
- Iterative Hessian sketch: fast and accurate solution approximation for constrained least-squares
- Lectures on Stochastic Programming
- Linear and nonlinear inverse problems with practical applications
- LSQR: An Algorithm for Sparse Linear Equations and Sparse Least Squares
- Nonstationary iterated Tikhonov regularization
- Numerical methods for coupled super-resolution
- On the choice of the regularization parameter for iterated Tikhonov regularization of ill-posed problems
- On variants of the Johnson–Lindenstrauss lemma
- On-Line Learning in Neural Networks
- Optimization methods for large-scale machine learning
- Paved with good intentions: analysis of a randomized block Kaczmarz method
- Randomized algorithms for estimating the trace of an implicit symmetric positive semi-definite matrix
- Randomized matrix-free trace and log-determinant estimators
- Semi-convergence properties of Kaczmarz's method
Cited in
(13)- Effective new methods for automated parameter selection in regularized inverse problems
- Hybrid projection methods with recycling for inverse problems
- Sampled limited memory methods for massive linear inverse problems
- Adaptive Tikhonov strategies for stochastic ensemble Kalman inversion
- slimTrain---A Stochastic Approximation Method for Training Separable Deep Neural Networks
- Hybrid Projection Methods with Recycling for Inverse Problems
- Non-contact detection of single-cell lead-acid battery electrodes’ defects through conductivity reconstruction by magnetic induction tomography
- An Optimal Scheduled Learning Rate for a Randomized Kaczmarz Algorithm
- A relaxed iterated Tikhonov regularization for linear ill-posed inverse problems
- A variable projection method for large-scale inverse problems with \(\ell^1\) regularization
- Solution of the EEG inverse problem by random dipole sampling
- Sampling linear inverse problems with noise
- Solving, tracking and stopping streaming linear inverse problems
This page was built for publication: Sampled Tikhonov regularization for large linear inverse problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4973539)
