NSPOD: Accelerating Krylov solvers via DeepONet-learned POD subspaces
Abstract
The convergence of Krylov-based linear iterative solvers applied to parametric partial differential equations (PDEs) is often highly sensitive to the domain, its discretization, the location/values of the applied Dirichlet/Neumann boundary conditions, body forces and material properties, among others. We have previously introduced hybridization of classical linear iterative solvers with neural operators for specific geometries, but they tend to not perform well on geometries not previously seen during training. We partially addressed this challenge by introducing the deep operator network Geo-DeepONet and hybridizing it with Krylov-based iterative linear solvers, which, despite learning effectively across arbitrary unstructured meshes without requiring retraining, led to only modest reductions in iterations compared to state-of-the-art preconditioners. In this study we introduce Neural Subspace Proper Orthogonal Decomposition (NSPOD), a multigrid-like deep operator network-based preconditioner which can dramatically reduce the number of iterations needed for convergence in Krylov-based linear iterative solvers, even when compared to state-of-the-art methods such as algebraic multigrid preconditioners. We demonstrate its efficiency via numerical experiments on a linearized version of solid mechanics PDEs applied to unstructured domains obtained from complex CAD geometries. We expect that the findings in this study lead to more efficient hybrid preconditioners that can match, or possibly even surpass, the convergence properties of the current gold standard preconditioning methods for solid mechanics PDEs.
Versions
Related
- Polynomial extrapolation techniques for accelerating Picard iterations with multigrid solvers for nonlinear problems in isogeometric analysis
- Stationary kinetic Krylov-Safonov estimates in space dimension one
- Factorized Krylov subspace methods for solving large Sylvester equations
- Comparative Analysis of Large-Scale SVM Solvers: Convergence, Resource Consumption, and Energy Footprint
- Impact of Feature Scaling on the Convergence Speed of Linear SVM Solvers : An Empirical Benchmark
BibTeX
@misc{levreroflorencio2026nspod,
title = {{NSPOD: Accelerating Krylov solvers via DeepONet-learned POD subspaces}},
author = {Francesc Levrero-Florencio and Youngkyu Lee and Jay Pathak and George Em Karniadakis},
howpublished = {arXiv},
year = {2026},
month = may,
doi = {10.48550/arxiv.2605.07828},
eprint = {2605.07828},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.07828},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - UNPB
TI - NSPOD: Accelerating Krylov solvers via DeepONet-learned POD subspaces
AU - Francesc Levrero-Florencio
AU - Youngkyu Lee
AU - Jay Pathak
AU - George Em Karniadakis
PY - 2026
DA - 2026/05/11
PB - arXiv
DO - 10.48550/arxiv.2605.07828
UR - https://arxiv.org/abs/2605.07828
AB - The convergence of Krylov-based linear iterative solvers applied to parametric partial differential equations (PDEs) is often highly sensitive to the domain, its discretization, the location/values of the applied Dirichlet/Neumann boundary conditions, body forces and material properties, among others. We have previously introduced hybridization of classical linear iterative solvers with neural operators for specific geometries, but they tend to not perform well on geometries not previously seen during training. We partially addressed this challenge by introducing the deep operator network Geo-DeepONet and hybridizing it with Krylov-based iterative linear solvers, which, despite learning effectively across arbitrary unstructured meshes without requiring retraining, led to only modest reductions in iterations compared to state-of-the-art preconditioners. In this study we introduce Neural Subspace Proper Orthogonal Decomposition (NSPOD), a multigrid-like deep operator network-based preconditioner which can dramatically reduce the number of iterations needed for convergence in Krylov-based linear iterative solvers, even when compared to state-of-the-art methods such as algebraic multigrid preconditioners. We demonstrate its efficiency via numerical experiments on a linearized version of solid mechanics PDEs applied to unstructured domains obtained from complex CAD geometries. We expect that the findings in this study lead to more efficient hybrid preconditioners that can match, or possibly even surpass, the convergence properties of the current gold standard preconditioning methods for solid mechanics PDEs.
ER -