Improved backward error bounds for LU and Cholesky factorizations

DOI10.1137/130927231MaRDI QIDQ2923365zbMATH OpenOpenAlexWikidataFDO

Authors Claude-Pierre Jeannerod, Siegfried M. Rump

Publication date 15 October 2014

Published in SIAM Journal on Matrix Analysis and Applications (Search for Journal in Brave)

Full work available at URL https://hal.inria.fr/hal-00841361/file/RumpJeannerod14.pdf

zbMATH Keywords

backward error boundCholesky factorizationfloating-point summationLU factorizationrounding error analysistriangular system solving

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Roundoff error (65G50)

Review text

New backward error bounds for LU and Cholesky factorizations are provided. To get these new bounds, the main ingradient is a refinement of the well-known Lemma 8.4 from \textit{N. J. Higham} [Accuracy and stability of numerical algorithms. 2nd ed. Philadelphia, PA: SIAM (2002; Zbl 1011.65010)] via some careful rounding error analysis of sums of one floating-point number and \(n-1\) real numbers.

Recommendations

• A new approach to backward error analysis of LU factorization
• Forward roundoff error analysis of the LU and Cholesky factorizations
• On the Perturbation of LU, Cholesky, and QR Factorizations
• A note on rounding-error analysis of Cholesky factorization

Cited in

(18)

• A new approach to backward error analysis of LU factorization
• A note on rounding-error analysis of Cholesky factorization
• Formal proofs of rounding error bounds. With application to an automatic positive definiteness check
• Error estimates for the summation of real numbers with application to floating-point summation
• Improved error bounds for floating-point products and Horner's scheme
• Floating-point arithmetic on the test bench. How are verified numerical solutions calculated?
• Forward roundoff error analysis of the LU and Cholesky factorizations
• On relative errors of floating-point operations: optimal bounds and applications
• Sharp estimates for perturbation errors in summations
• Mathematically rigorous global optimization in floating-point arithmetic
• A New Approach to Probabilistic Rounding Error Analysis
• A radix-independent error analysis of the Cornea-Harrison-Tang method
• Roundoff-Error-Free Basis Updates of LU Factorizations for the Efficient Validation of Optimality Certificates
• Improved rigorous perturbation bounds for the LU and QR factorizations.
• Stochastic rounding and its probabilistic backward error analysis
• Mixed and componentwise condition numbers for matrix decompositions
• Analyzing vector orthogonalization algorithms
• An improved shifted CholeskyQR based on columns

This page was built for publication: Improved backward error bounds for LU and Cholesky factorizations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2923365)