An incomplete Cholesky factorization for dense symmetric positive definite matrices

DOI10.1023/A:1022323931043MaRDI QIDQ1587315zbMATH OpenFDO

Authors Chih-Jen Lin, Romesh Saigal

Publication date 3 May 2001

Published in BIT (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1023/A:1022323931043

zbMATH Keywords

Cholesky factorizationdense matricesincomplete Cholesky decompositionnumerical results

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Iterative numerical methods for linear systems (65F10) Numerical computation of matrix norms, conditioning, scaling (65F35)

Review text

An incomplete Cholesky factorization is obtained if only the large components of pivot columns enter into the factorization. The efficiency is tested for three types of matrices. 1. Randomly generated matrices; 2. Dense matrices from semi-definite programming of the form \(\bar A (S^{-1} \otimes X) A^T\) and 3. Dense matrices from support vector machines. Numerical results are listed in 9 tables.

Recommendations

• Parallel Processing and Applied Mathematics
• Incomplete Cholesky Factorizations with Limited Memory
• MODIFLED INCOMPLETE CHOLESKY FACTORIZATION PRECONDITIONERS FOR A SYMMETRIC POSITIVE DEFINITE MATRIX
• Low-density incomplete \(LDL^T\) factorizations
• On the incomplete Cholesky decomposition of a class of perturbed matrices

Cited in

(16)

• Preconditioning techniques for large linear systems: A survey
• Convex optimization techniques in compliant assembly simulation
• On partial Cholesky factorization and a variant of quasi-Newton preconditioners for symmetric positive definite matrices
• Partitioned solver for strongly coupled fluid-structure interaction
• Low-density incomplete \(LDL^T\) factorizations
• A projected gradient algorithm for solving the maxcut SDP relaxation
• On the incomplete Cholesky decomposition of a class of perturbed matrices
• Robust Approximate Cholesky Factorization of Rank-Structured Symmetric Positive Definite Matrices
• The Discrete Empirical Interpolation Method: Canonical Structure and Formulation in Weighted Inner Product Spaces
• A kernel-based core growing clustering method
• Incomplete Cholesky Factorizations with Limited Memory
• On the stability of the incomplete Cholesky decomposition for a singular perturbed problem, where the coefficient matrix is not an M‐matrix
• Parallel Processing and Applied Mathematics
• An efficient computation of the inverse of the single layer matrix for the resolution of the linear elasticity problem in BEM
• Cholesky decomposition of a positive semidefinite matrix with known kernel
• On the solution of large-scale SDP problems by the modified barrier method using iterative solvers

Describes a project that uses

Uses Software

• QAPLIB
• ITSOL
• SVMlight
• SPARSKIT

This page was built for publication: An incomplete Cholesky factorization for dense symmetric positive definite matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1587315)