Automated Grammar-based Algebraic Multigrid Design With Evolutionary Algorithms

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parthasarathy, Dinesh, Mitchell, Wayne, Gambhir, Arjun, Köstler, Harald, Rüde, Ulrich
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908897453801472
author Parthasarathy, Dinesh
Mitchell, Wayne
Gambhir, Arjun
Köstler, Harald
Rüde, Ulrich
author_facet Parthasarathy, Dinesh
Mitchell, Wayne
Gambhir, Arjun
Köstler, Harald
Rüde, Ulrich
contents Although multigrid is asymptotically optimal for solving many important partial differential equations, its efficiency relies heavily on the careful selection of the individual algorithmic components. In contrast to recent approaches that can optimize certain multigrid components using deep learning techniques, we adopt a complementary strategy, employing evolutionary algorithms to construct efficient multigrid cycles from proven algorithmic building blocks. Here, we will present its application to generate efficient algebraic multigrid methods with so-called \emph{flexible cycling}, that is, level-specific smoothing sequences and non-recursive cycling patterns. The search space with such non-standard cycles is intractable to navigate manually, and is generated using genetic programming (GP) guided by context-free grammars. Numerical experiments with the linear algebra library, \emph{hypre}, demonstrate the potential of these non-standard GP cycles to improve multigrid performance both as a solver and a preconditioner.
format Preprint
id arxiv_https___arxiv_org_abs_2603_17641
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Automated Grammar-based Algebraic Multigrid Design With Evolutionary Algorithms
Parthasarathy, Dinesh
Mitchell, Wayne
Gambhir, Arjun
Köstler, Harald
Rüde, Ulrich
Computational Engineering, Finance, and Science
Artificial Intelligence
Numerical Analysis
65M55, 68W50, 35-04
G.1.8; I.2.2; F.4.3
Although multigrid is asymptotically optimal for solving many important partial differential equations, its efficiency relies heavily on the careful selection of the individual algorithmic components. In contrast to recent approaches that can optimize certain multigrid components using deep learning techniques, we adopt a complementary strategy, employing evolutionary algorithms to construct efficient multigrid cycles from proven algorithmic building blocks. Here, we will present its application to generate efficient algebraic multigrid methods with so-called \emph{flexible cycling}, that is, level-specific smoothing sequences and non-recursive cycling patterns. The search space with such non-standard cycles is intractable to navigate manually, and is generated using genetic programming (GP) guided by context-free grammars. Numerical experiments with the linear algebra library, \emph{hypre}, demonstrate the potential of these non-standard GP cycles to improve multigrid performance both as a solver and a preconditioner.
title Automated Grammar-based Algebraic Multigrid Design With Evolutionary Algorithms
topic Computational Engineering, Finance, and Science
Artificial Intelligence
Numerical Analysis
65M55, 68W50, 35-04
G.1.8; I.2.2; F.4.3
url https://arxiv.org/abs/2603.17641