When can a neural operator replace a coarse solve? Architectural principles for two-level preconditioning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Melchers, Hugo, Abdelmalik, Michael, Dolean, Victorita
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916028066299904
author Melchers, Hugo
Abdelmalik, Michael
Dolean, Victorita
author_facet Melchers, Hugo
Abdelmalik, Michael
Dolean, Victorita
contents Neural operators are increasingly used as drop-in accelerators inside classical numerical methods, but it is rarely clear which architectural ingredients matter for which role. We answer this question for one important role: the coarse-space correction inside a two-level preconditioner for discretised linear partial differential equations. By systematically varying four DeepONet-like architectures along two design axes - input discretisation (sampling versus integration against a basis) and source-term linearity - we show that the favourable corner of this 2$\times$2 design is occupied by a single architecture, the Neural Green's Operator (NGO), and that moving away from it produces predictable failure modes: structurally non-symmetric preconditioned spectra, breakdown of preconditioned conjugate gradients on self-adjoint problems, and stagnation on non-self-adjoint ones. Used as a coarse-space correction, the NGO matches the iteration count of an exact coarse solve on diffusion and advection-diffusion problems. We also characterise the failure of fixed-size learned coarse spaces at high Helmholtz wave numbers, isolating it as a property of the basis rather than of the architecture. The principle generalises: integrating inputs against the basis used for the output is what allows a neural operator to serve as a Galerkin-type coarse-space correction.
format Preprint
id arxiv_https___arxiv_org_abs_2605_19867
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle When can a neural operator replace a coarse solve? Architectural principles for two-level preconditioning
Melchers, Hugo
Abdelmalik, Michael
Dolean, Victorita
Numerical Analysis
68T07 (Primary) 65F08, 65N55 (Secondary)
I.2.6; G.1.3; G.1.8
Neural operators are increasingly used as drop-in accelerators inside classical numerical methods, but it is rarely clear which architectural ingredients matter for which role. We answer this question for one important role: the coarse-space correction inside a two-level preconditioner for discretised linear partial differential equations. By systematically varying four DeepONet-like architectures along two design axes - input discretisation (sampling versus integration against a basis) and source-term linearity - we show that the favourable corner of this 2$\times$2 design is occupied by a single architecture, the Neural Green's Operator (NGO), and that moving away from it produces predictable failure modes: structurally non-symmetric preconditioned spectra, breakdown of preconditioned conjugate gradients on self-adjoint problems, and stagnation on non-self-adjoint ones. Used as a coarse-space correction, the NGO matches the iteration count of an exact coarse solve on diffusion and advection-diffusion problems. We also characterise the failure of fixed-size learned coarse spaces at high Helmholtz wave numbers, isolating it as a property of the basis rather than of the architecture. The principle generalises: integrating inputs against the basis used for the output is what allows a neural operator to serve as a Galerkin-type coarse-space correction.
title When can a neural operator replace a coarse solve? Architectural principles for two-level preconditioning
topic Numerical Analysis
68T07 (Primary) 65F08, 65N55 (Secondary)
I.2.6; G.1.3; G.1.8
url https://arxiv.org/abs/2605.19867