Efficient solution of parameter identification problems with H¹ regularization

DOI10.1137/22M1520591arXiv2209.02815OpenAlexW4393422832MaRDI QIDQ6409906FDOQ6409906

Authors: Jan Blechta, Oliver G. Ernst 

Publication date: 6 September 2022

Abstract: We consider the identification of spatially distributed parameters under $H^1$ regularization. Solving the associated minimization problem by Gauss-Newton iteration results in linearized problems to be solved in each step that can be cast as boundary value problems involving a low-rank modification of the Laplacian. Using algebraic multigrid as a fast Laplace solver, the Sherman-Morrison-Woodbury formula can be employed to construct a preconditioner for these linear problems which exhibits excellent scaling w.r.t. the relevant problem parameters. We first develop this approach in the functional setting, thus obtaining a consistent methodology for selecting boundary conditions that arise from the $H^1$ regularization. We then construct a method for solving the discrete linear systems based on combining any fast Poisson solver with the Woodbury formula. The efficacy of this method is then demonstrated with scaling experiments. These are carried out for a common nonlinear parameter identification problem arising in electrical resistivity tomography.

Full work available at URL: https://doi.org/10.1137/22m1520591

Recommendations

• A fast iterative method for solving regularized parameter identification problems in elliptic boundary value problems
• An efficient linear solver for nonlinear parameter identification problems
• Multigrid solution of a linearized, regularized least-squares problem in electrical impedance tomography
• An overlapping domain decomposition method for parameter identification problems
• On level set regularization for highly ill-posed distributed parameter estimation problems

Mathematics Subject Classification ID

Numerical solutions to overdetermined systems, pseudoinverses (65F20) Preconditioners for iterative methods (65F08) Iterative numerical methods for linear systems (65F10) Numerical computation of matrix norms, conditioning, scaling (65F35) Numerical solutions to equations with linear operators (65J10)

Cites Work

• Gmsh: a 3-D finite element mesh generator with built-in pre- and post-processing facilities
• HSL_MI20 : An efficient AMG preconditioner for finite element problems in 3D
• Title not available (Why is that?)
• A Robust Preconditioner with Low Memory Requirements for Large Sparse Least Squares Problems
• Computational Methods for Inverse Problems
• Inverse acoustic and electromagnetic scattering theory
• Continuity of Solutions of Parabolic and Elliptic Equations
• Mixed finite element methods and applications
• Numerical methods in matrix computations
• A note on preconditioning nonsymmetric matrices
• Solution of Sparse Indefinite Systems of Linear Equations
• A Note on Preconditioning for Indefinite Linear Systems
• Title not available (Why is that?)
• Preconditioning discretizations of systems of partial differential equations.
• Geometric aspects of the theory of Krylov subspace methods
• Finite elements and fast iterative solvers. With applications in incompressible fluid dynamics
• Optimal Preconditioning for Raviart--Thomas Mixed Formulation of Second-Order Elliptic Problems
• Preconditioned iterative methods for solving linear least squares problems
• Inner-Iteration Krylov Subspace Methods for Least Squares Problems
• Krylov subspace methods. Principles and analysis.
• Title not available (Why is that?)
• A Note on the Superlinear Convergence of GMRES
• Block triangular preconditioners for nonsymmetric saddle point problems: Field-of-values analysis
• Some Superlinear Convergence Results for the Conjugate Gradient Method
• Stability of Conjugate Gradient and Lanczos Methods for Linear Least Squares Problems
• Computational methods in geophysical electromagnetics
• Subspace preconditioned LSQR for discrete ill-posed problems
• Two-level preconditioners for regularized inverse problems. I: Theory
• The State-of-the-Art of Preconditioners for Sparse Linear Least-Squares Problems
• An improved preconditioned LSQR for discrete ill-posed problems
• Iterative solution of symmetric quasi-definite linear systems
• A hybrid LSMR algorithm for large-scale Tikhonov regularization
• Preconditioner based on the Sherman-Morrison formula for regularized least squares problems
• Preconditioners for rank deficient least squares problems
• Preconditioning linear least-squares problems by identifying a basis matrix
• On Using Cholesky-Based Factorizations and Regularization for Solving Rank-Deficient Sparse Linear Least-Squares Problems
• Bayes meets Krylov: statistically inspired preconditioners for CGLS
• Superlinear convergence of Krylov subspace methods for self-adjoint problems in Hilbert space
• Spectral method approximation of flow optimal control problems with \(H^1\)-norm state constraint
• Priorconditioners for linear systems
• Preconditioning of linear least squares by robust incomplete factorization for implicitly held normal equations
• A fast solver for an H1 regularized PDE-constrained optimization problem
• Stability of linear GMRES convergence with respect to compact perturbations
• Convergence analysis of LSQR for compact operator equations
• Large-Scale Inverse Problems in Imaging

This page was built for publication: Efficient solution of parameter identification problems with $H^1$ regularization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6409906)