What question did this study set out to answer?

This research aims to develop robust and efficient methods for solving large sparse linear least-squares problems using mixed precision techniques.

August 2, 2026ACM Transactions on Mathematical Software

A computational study of low precision incomplete Cholesky factorization preconditioners for sparse linear least-squares problems

View Full Paper

Ask AI

Bookmark

Authors

JSJ. A. Scott MTMiroslav Tůma

Discussion

Member takes

Overview

Randomized trial explores mixed precision approaches to enhance solution quality in sparse linear least-squares problems, suggesting practical applications.

Key Points

This research aims to develop robust and efficient methods for solving large sparse linear least-squares problems using mixed precision techniques.
Utilized low precision incomplete Cholesky factorization preconditioners in solving least-squares problems.
Conducted experiments using problems derived from practical applications to assess performance.
Examined both level-based and memory-limited preconditioners for their effectiveness.
Level-based preconditioners were found to be ineffective for least-squares problems.
Memory-limited preconditioners yielded high-quality solutions; half precision can be adequate when high accuracy is not critical.
Single precision can minimize memory use while still recovering double precision accuracy, even for ill-conditioned problems.

Cite This Study

Scott et al. (2026) studied this question.

synapsesocial.com/papers/6a6eeb201b0468a7eeab4024 https://doi.org/10.1145/3837072

View Full Paper

Ask AI

Bookmark

August 2, 2026ACM Transactions on Mathematical Software

A computational study of low precision incomplete Cholesky factorization preconditioners for sparse linear least-squares problems

View Full Paper

Ask AI

Bookmark

Authors

JSJ. A. Scott MTMiroslav Tůma

Discussion

Member takes

Overview

Randomized trial explores mixed precision approaches to enhance solution quality in sparse linear least-squares problems, suggesting practical applications.

Key Points

This research aims to develop robust and efficient methods for solving large sparse linear least-squares problems using mixed precision techniques.
Utilized low precision incomplete Cholesky factorization preconditioners in solving least-squares problems.
Conducted experiments using problems derived from practical applications to assess performance.
Examined both level-based and memory-limited preconditioners for their effectiveness.
Level-based preconditioners were found to be ineffective for least-squares problems.
Memory-limited preconditioners yielded high-quality solutions; half precision can be adequate when high accuracy is not critical.
Single precision can minimize memory use while still recovering double precision accuracy, even for ill-conditioned problems.

Cite This Study

Scott et al. (2026) studied this question.

synapsesocial.com/papers/6a6eeb201b0468a7eeab4024 https://doi.org/10.1145/3837072

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark