Preconditioned conjugate gradient method for rank deficient least-squares problems

DOI10.1080/00207169908804872MaRDI QIDQ4944666zbMATH OpenOpenAlexFDO

Authors C. H. Santos, Jinyun Yuan

Publication date 13 November 2000

Published in International Journal of Computer Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/00207169908804872

zbMATH Keywords

comparison of methodsconjugate gradient methoderror boundsiterative methodlarge sparse problemspreconditioningrank deficient least squares problemssuccessive overrelaxation

Mathematics Subject Classification ID

Iterative numerical methods for linear systems (65F10) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Numerical computation of matrix norms, conditioning, scaling (65F35) Computational methods for sparse matrices (65F50)

Review text

Some conjugate gradient methods for solving rank deficient least squares problems are proposed. Such iterative methods are preferred for handling large sparse problems. Error bounds and comparison with 2-block and 3-block successive overrelaxation methods are investigated. It is shown theoretically that the best iterative method is the conjugate gradient method for solving rank deficient least squares problems. No results of computational experiments are given.

Recommendations

• Preconditioners for rank deficient least squares problems
• Conjugate gradient method for rank deficient saddle point problems
• Study of 2-block AOR iterative method for rank deficient least squares problems
• Block SOR methods for rank-deficient least-squares problems
• scientific article; zbMATH DE number 1507112

Cites work

• An implicit deflation method for ill-conditioned Sylvester and Lyapunov equations
• Block SOR methods for rank-deficient least-squares problems
• Convergence of a direct-iterative method for large-scale least-squares problems
• On the rate of convergence of the preconditioned conjugate gradient method
• Successive overrelaxation methods for solving the rank deficient linear least squares problem

Cited in

(16)

• Block SOR methods for rank-deficient least-squares problems
• Rank-deficient prewhitening with quotient SVD and ULV decompositions
• Off-diagonal low-rank preconditioner for difficult PageRank problems
• Preconditioners for rank deficient least squares problems
• Optimal parameters of GSOR-like methods for solving the augmented linear systems
• Symmetric-triangular decomposition and its applications. II: Preconditioners for indefinite systems
• The direct-projection methods
• Preconditioners for the conjugate gradient algorithm using Gram–Schmidt and least squares methods
• scientific article; zbMATH DE number 1507112 (Why is no real title available?)
• Conjugate gradient method for rank deficient saddle point problems
• A modified symmetric successive overrelaxation method for augmented systems
• Evaluation of ST preconditioners for saddle point problems
• Symmetric SOR method for augmented systems
• A class of triangular splitting methods for saddle point problems
• Generalization: One technique of computational and applied mathematical methodology
• On generalized symmetric SOR method for augmented systems

This page was built for publication: Preconditioned conjugate gradient method for rank deficient least-squares problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4944666)