Learning Algebraic Multigrid Using Graph Neural Networks

03/12/2020
by   Ilay Luz, et al.
0

Efficient numerical solvers for sparse linear systems are crucial in science and engineering. One of the fastest methods for solving large-scale sparse linear systems is algebraic multigrid (AMG). The main challenge in the construction of AMG algorithms is the selection of the prolongation operator – a problem-dependent sparse matrix which governs the multiscale hierarchy of the solver and is critical to its efficiency. Over many years, numerous methods have been developed for this task, and yet there is no known single right answer except in very special cases. Here we propose a framework for learning AMG prolongation operators for linear systems with sparse symmetric positive (semi-) definite matrices. We train a single graph neural network to learn a mapping from an entire class of such matrices to prolongation operators, using an efficient unsupervised loss function. Experiments on a broad class of problems demonstrate improved convergence rates compared to classical AMG, demonstrating the potential utility of neural networks for developing sparse system solvers.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/25/2019

Learning to Optimize Multigrid PDE Solvers

Constructing fast numerical solvers for partial differential equations (...
research
03/14/2022

Towards Neural Sparse Linear Solvers

Large sparse symmetric linear systems appear in several branches of scie...
research
06/03/2021

Optimization-Based Algebraic Multigrid Coarsening Using Reinforcement Learning

Large sparse linear systems of equations are ubiquitous in science and e...
research
07/19/2023

AutoAMG(θ): An Auto-tuned AMG Method Based on Deep Learning for Strong Threshold

Algebraic Multigrid (AMG) is one of the most widely used iterative algor...
research
12/16/2022

Generalizing Reduction-Based Algebraic Multigrid

Algebraic Multigrid (AMG) methods are often robust and effective solvers...
research
06/29/2020

AMG preconditioners for Linear Solvers towards Extreme Scale

Linear solvers for large and sparse systems are a key element of scienti...
research
06/27/2019

Fast Training of Sparse Graph Neural Networks on Dense Hardware

Graph neural networks have become increasingly popular in recent years d...

Please sign up or login with your details

Forgot password? Click here to reset