Automated tuning for the parameters of linear solvers

DOI10.1016/j.jcp.2023.112533arXiv2303.15451OpenAlexW4387266500MaRDI QIDQ6087949

Boris I. Krasnopolsky, Andrey Petrushov

Publication date: 16 November 2023

Published in: Journal of Computational Physics (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2303.15451

zbMATH Keywords

algebraic multigrid methodmachine learningsystems of linear algebraic equationsparameters optimizationhybrid evolution strategy

Mathematics Subject Classification ID

Numerical linear algebra (65Fxx) Artificial intelligence (68Txx) Computer aspects of numerical algorithms (65Yxx)

Related Items (2)

Tuning soft mutations of the evolution algorithm for optimizing the linear solver parameters ⋮ An approach to the implementation of the multigrid method with full approximation for CFD problems

Cites Work

• Unnamed Item
• Evolution strategies. A comprehensive introduction
• An investigation for loss functions widely used in machine learning
• Finite difference method for incompressible Navier--Stokes equations in arbitrary orthogonal curvilinear coordinates
• Reducing Parallel Communication in Algebraic Multigrid through Sparsification
• Performance-Based Numerical Solver Selection in the Lighthouse Framework
• The university of Florida sparse matrix collection
• On long-range interpolation operators for aggressive coarsening
• Bi-CGSTAB: A Fast and Smoothly Converging Variant of Bi-CG for the Solution of Nonsymmetric Linear Systems
• Introduction to Genetic Algorithms
• Third‐order‐accurate semi‐implicit Runge–Kutta scheme for incompressible Navier–Stokes equations
• Automated empirical optimizations of software and the ATLAS project

This page was built for publication: Automated tuning for the parameters of linear solvers