Error analysis of matrix multiplication with narrow range floating-point arithmetic

DOI10.1137/24M1685109MaRDI QIDQ6959698zbMATH OpenFDO

Authors Theo A. Mary, Mantas Mikaitis

Publication date 15 July 2025

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/24M1685109

zbMATH Keywords

floating-point arithmeticGPU tensor coreslow precisionmatrix multiplicationrounding error analysisunderflow

Mathematics Subject Classification ID

Numerical linear algebra (65F99) Roundoff error (65G50) Numerical algorithms for computer arithmetic, etc. (65Y04) Numerical algorithms for specific classes of architectures (65Y10)

Cites work

• Accuracy and Stability of Numerical Algorithms
• CPFloat: a C library for simulating low-precision arithmetic
• Improved error bounds for inner products in floating-point arithmetic
• Matrix Multiplication in Multiword Arithmetic: Error Analysis and Application to GPU Tensor Cores
• Mixed precision block fused multiply-add: error analysis and application to GPU tensor cores
• Squeezing a Matrix into Half Precision, with an Application to Solving Linear Systems
• Underflow and the Reliability of Numerical Software

This page was built for publication: Error analysis of matrix multiplication with narrow range floating-point arithmetic

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6959698)