Truncated QR Factorization with Pivoting in Mixed Precision
SIAM Journal on Scientific Computing · 30.04.2025 · 47 (2) · B382-B401
DOI: 10.1137/24m1644705 ↗Kurz gefasst
Fachartikel in der Zeitschrift SIAM Journal on Scientific Computing, online erschienen am 12.03.2025. Verfasst von Alfredo Buttari und 2 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 25 Literaturangaben. Gefördert von: Agence Nationale de la Recherche, National Polytechnic Institute of Toulouse.
Abstract (Originalsprache)
Low-rank approximations are widely used to reduce the memory footprint and operational complexity of numerous linear algebra algorithms in scientific computing and data analysis. In some of our recent work we have demonstrated that low-rank approximations can be stored using multiple arithmetic precisions to further reduce the storage and execution time. In this work we present a method that can produce this mixed-precision representation directly; this relies on a mixed-precision truncated rank-revealing QR factorization with pivoting. We present a floating-point error analysis and provide bounds on the error of the approximation demonstrating that the use of multiple precisions does not alter the overall accuracy. Finally, we present experimental results showing the execution time reduction for the cases where either classical or randomized pivoting are used. Reproducibility of computational results. This paper has been awarded the “SIAM Reproducibility Badge: Code and data available” as a recognition that the authors have followed reproducibility principles valued by SISC and the scientific computing community. Code and data that allow readers to reproduce the results in this paper are available at https://gitlab.com/mpqr/mpqr and in the supplementary materials ( mpqr-main.zip [334KB]). [Formula: see text]
Förderung
- Agence Nationale de la Recherche
- National Polytechnic Institute of Toulouse
Zitieren
APA
Buttari, A., Mary, T., & Pacteau, A. (2025). Truncated QR Factorization with Pivoting in Mixed Precision. SIAM Journal on Scientific Computing, 47(2), B382-B401. https://doi.org/10.1137/24m1644705
Vancouver
Buttari A, Mary T, Pacteau A. Truncated QR Factorization with Pivoting in Mixed Precision. SIAM Journal on Scientific Computing. 2025;47(2):B382-B401. doi:10.1137/24m1644705
BibTeX
@article{Buttari2025,
author = {Buttari, Alfredo and Mary, Theo and Pacteau, André},
title = {Truncated QR Factorization with Pivoting in Mixed Precision},
journal = {SIAM Journal on Scientific Computing},
year = {2025},
volume = {47},
number = {2},
pages = {B382--B401},
doi = {10.1137/24m1644705}
}
EUVIMED