NIST - National Institute of Standards and Technology

07/27/2026 | Press release | Distributed by Public on 07/28/2026 03:08

librla: A library of randomized linear algebra routines

Published
July 27, 2026

Author(s)

Zydrunas Gimbutas, Adrianna Gillman

Abstract

Randomized linear algebra algorithms have become a vital tool for a variety of areas including fast direct solvers, reduced order modeling, and data science. Additionally, randomized linear algebra provides useful tools for solving total least squares problems, rank deficient least squares problem, doing matrix approximation, and skeletonizing (i.e. subset selection) a matrix. librla provides low rank QR factorizations, SVDs and interpolatory decompositions written natively in Python, Julia and Matlab. The algorithms randomly sample the range of the matrix or operator in a similar manner to (Halko et al., 2011). A key feature of this package is that it is designed to exploit Level 3 BLAS operators as much as possible. librla is designed for small to mid-range sized matrices (i.e. up to roughly 10,000 in size depending on computing resources).
Citation
(*TBD - To Be Determined*)
Pub Type
Journals

Keywords

Randomized methods, QR factorization, singular value decomposition, interpolatory decomposition, Python, Matlab, Octave, Julia

Citation

Gimbutas, Z. and Gillman, A. (2026), librla: A library of randomized linear algebra routines, (*TBD - To Be Determined*) (Accessed July 28, 2026)
Additional citation formats

Issues

If you have any questions about this publication or are having problems accessing it, please contact [email protected].

NIST - National Institute of Standards and Technology published this content on July 27, 2026, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on July 28, 2026 at 09:08 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]