Numerical stability of algorithms at extreme scale and low precisions

DOI10.4171/icm2022/74MaRDI QIDQ6200205

Nicholas J. Higham

Publication date: 22 March 2024

Published in: International Congress of Mathematicians (Search for Journal in Brave)

zbMATH Keywords

numerical stabilityfloating-point arithmeticbackward error analysisblocked algorithmfused multiply-addmixed precision computationprobabilistic rounding error analysis

Mathematics Subject Classification ID

Roundoff error (65G50) Direct numerical methods for linear systems and matrix inversion (65F05)

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Reducing Floating Point Error in Dot Product Using the Superblock Family of Algorithms
• Stochastic Perturbation Theory
• Mixed Precision Block Fused Multiply-Add: Error Analysis and Application to GPU Tensor Cores
• Anatomy of high-performance matrix multiplication
• LAPACK Users' Guide
• Error Analysis of Direct Methods of Matrix Inversion
• A set of level 3 basic linear algebra subprograms
• Handbook of Floating-Point Arithmetic
• Accelerating the Solution of Linear Systems by Iterative Refinement in Three Precisions
• Accuracy and Stability of Numerical Algorithms
• A Class of Fast and Accurate Summation Algorithms
• Sharper Probabilistic Backward Error Analysis for Basic Linear Algebra Kernels with Random Data
• Probabilistic Error Analysis for Inner Products
• Random Matrices Generating Large Growth in LU Factorization with Pivoting
• Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems
• Squeezing a Matrix into Half Precision, with an Application to Solving Linear Systems
• A New Approach to Probabilistic Rounding Error Analysis
• Modern Error Analysis
• Stochastic Rounding and Its Probabilistic Backward Error Analysis