Improving the Numerical Stability of Fast Matrix Multiplication

DOI10.1137/15M1032168zbMath1348.65080arXiv1507.00687MaRDI QIDQ2827068

Alex Druinsky, Benjamin Lipshitz, Oded Schwartz, Austin R. Benson, Grey Ballard

Publication date: 12 October 2016

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

Full work available at URL: https://arxiv.org/abs/1507.00687

zbMATH Keywords

error boundsdiagonal scalingpractical fast matrix multiplication

Mathematics Subject Classification ID

Roundoff error (65G50) Numerical computation of matrix norms, conditioning, scaling (65F35)

Related Items (5)

Pebbling Game and Alternative Basis for High Performance Matrix Multiplication ⋮ Numerical stability and tensor nuclear norm ⋮ Fast matrix multiplication and its algebraic neighbourhood ⋮ Strassen's Algorithm for Tensor Contraction ⋮ Equivalent polyadic decompositions of matrix multiplication tensors

Uses Software

• mctoolbox
• z3

Cites Work

• Unnamed Item
• The aggregation and cancellation techniques as a practical tool for faster matrix multiplication
• Fast matrix multiplication is stable
• Stability of fast algorithms for matrix multiplication
• On practical algorithms for accelerated matrix multiplication
• Improving and estimating the accuracy of Strassen's algorithm
• On the complexity of the multiplication of matrices of small formats
• Fast linear algebra is stable
• Gaussian elimination is not optimal
• Error analysis of algorithms for matrix multiplication and triangular decomposition using Winograd's identity
• The bilinear complexity and practical algorithms for matrix multiplication
• Powers of tensors and fast matrix multiplication
• Noncommutative Bilinear Algorithms for $3 \times 3$ Matrix Multiplication
• The vec-permutation matrix, the vec operator and Kronecker products: a review
• Computational Complexity and Numerical Stability
• Accuracy and Stability of Numerical Algorithms
• Graph Expansion Analysis for Communication Costs of Fast Rectangular Matrix Multiplication
• Graph expansion and communication costs of fast matrix multiplication
• New Lower Bounds for the Rank of Matrix Multiplication

This page was built for publication: Improving the Numerical Stability of Fast Matrix Multiplication