Low-rank approximate inverse for preconditioning tensor-structured linear systems

DOI10.1137/130918137MaRDI QIDQ2930013OpenAlexarXivFDO

Authors A. Nouy, L. Giraldi, G. Legrain

Publication date 17 November 2014

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

Full work available at URL https://arxiv.org/abs/1304.6004

zbMATH Keywords

approximate inversegreedy algorithmhierarchical Tucker tensorhigh dimensionlinear systemlow ranknumerical experimentPoisson equationpreconditioningstochastic partial differential equationtensor format

Mathematics Subject Classification ID

Theory of matrix inversion and generalized inverses (15A09) Multilinear algebra, tensor calculus (15A69) Laplace operator, Helmholtz equation (reduced wave equation), Poisson equation (35J05) Stochastic partial differential equations (aspects of stochastic analysis) (60H15) Numerical solutions to stochastic differential and integral equations (65C30) Preconditioners for iterative methods (65F08) Iterative numerical methods for linear systems (65F10) Computational methods for sparse matrices (65F50)

Abstract: In this paper, we propose an algorithm for the construction of low-rank approximations of the inverse of an operator given in low-rank tensor format. The construction relies on an updated greedy algorithm for the minimization of a suitable distance to the inverse operator. It provides a sequence of approximations that are defined as the projections of the inverse operator in an increasing sequence of linear subspaces of operators. These subspaces are obtained by the tensorization of bases of operators that are constructed from successive rank-one corrections. In order to handle high-order tensors, approximate projections are computed in low-rank Hierarchical Tucker subsets of the successive subspaces of operators. Some desired properties such as symmetry or sparsity can be imposed on the approximate inverse operator during the correction step, where an optimal rank-one correction is searched as the tensor product of operators with the desired properties. Numerical examples illustrate the ability of this algorithm to provide efficient preconditioners for linear systems in tensor format that improve the convergence of iterative solvers and also the quality of the resulting low-rank approximations of the solution.

Recommendations

• Preconditioned low-rank Riemannian optimization for linear systems with tensor product structure
• Tensor-structured preconditioners and approximate inverse of elliptic operators in \(\mathbb R^{d}\)
• Greedy low-rank approximation in Tucker format of solutions of tensor linear systems
• Krylov subspace methods for linear systems with tensor product structure
• Stability of low-rank tensor representations and structured multilevel preconditioning for elliptic PDEs

Cited in

(8)

• Greedy low-rank approximation in Tucker format of solutions of tensor linear systems
• Stochastic Galerkin methods for the steady-state Navier-Stokes equations
• Interpolation of inverse operators for preconditioning parameter-dependent equations
• Regularized computation of approximate pseudoinverse of large matrices using low-rank tensor train decompositions
• Greedy algorithms for high-dimensional eigenvalue problems
• Weakly Intrusive Low-Rank Approximation Method for Nonlinear Parameter-Dependent Equations
• Preconditioned low-rank Riemannian optimization for linear systems with tensor product structure
• Tensor-structured preconditioners and approximate inverse of elliptic operators in \(\mathbb R^{d}\)

This page was built for publication: Low-rank approximate inverse for preconditioning tensor-structured linear systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2930013)