21 December 2016 in Numerical Linear Algebra with Applications
1 May 2019 in Journal of Computational Physics
1 January 2024 in Communications in Computational Physics
Algebraic Multigrid (AMG) is one of the most widely used iterative algorithms for solving large sparse linear equations $Ax=b.$ In AMG, the coarse grid is a key component that affects the efficiency of the algorithm, the construction of which relies on the strong threshold parameter $θ.$ This parameter is generally chosen empirically, with a default value in many current AMG solvers of 0.25 for 2D problems and 0.5 for 3D problems. However, for many practical problems, the quality of the coarse grid and the efficiency of the AMG algorithm are sensitive to $θ;$ the default value is rarely optimal, and sometimes is far from it. Therefore, how to choose a better $θ$ is an important question. In this paper, we propose a deep learning based auto-tuning method, AutoAMG($θ$) for multiscale sparse linear equations, which are common in practical problems. The method uses Graph Neural Network (GNN) to extract matrix features, and a Multilayer Perceptron (MLP) to build the mapping between matrix features and the optimal $θ,$ which can adaptively predict $θ$ values for different matrices. Numerical experiments show that AutoAMG($θ$) can achieve significant speedup compared to the default $θ$ value. AMG, strong threshold, graph neural network, auto-tuning, multiscale matrix. 65F08, 65F10, 65N55, 68T05 COPYRIGHT: © Global Science Press
1 January 2015 in Communications in Computational Physics
An Adaptive Combined Preconditioner with Applications in Radiation Diffusion Equations - Volume 18 Issue 5 - Xiaoqiang Yue, Shi Shu, Xiao wen Xu, Zhiyang Zhou
25 October 2022 in CCF Transactions on High Performance Computing
11 May 2020 in CCF Transactions on High Performance Computing