EUVIMEDEUVIMED entdecken
Crossref

Automatic discovery of optimal meta-solvers via multi-objective optimization

Youngkyu Lee, Shanqing Liu, Jérôme Darbon, George Em Karniadakis

Journal of Computational Physics · 2026-10 · 563 · 115015

DOI: 10.1016/j.jcp.2026.115015 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computational Physics, 2026-10. Verfasst von Youngkyu Lee und 3 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 38 Literaturangaben. Gefördert von: Defense Advanced Research Projects Agency.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We design two classes of ultra-fast meta-solvers for linear systems arising after discretizing PDEs by combining neural operators with either simple iterative solvers, e.g., Jacobi and Gauss-Seidel, or with Krylov methods, e.g., GMRES and BiCGStab, using the trunk basis of DeepONet as a coarse preconditioner. The idea is to leverage the spectral bias of neural networks to account for the lower part of the spectrum in the error distribution while the upper part is handled easily and inexpensively using relaxation methods or fine-scale preconditioners. We create a pareto front of optimal meta-solvers using a plurarilty of metrics, and we introduce a preference function to select the best solver most suitable for a specific scenario. This automation for finding optimal solvers can be extended to nonlinear systems and other setups, e.g. finding the best meta-solver for space-time in time-dependent PDEs.

Förderung

  • Defense Advanced Research Projects Agency

Zitieren

APA

Lee, Y., Liu, S., Darbon, J., & Karniadakis, G. E. (2026). Automatic discovery of optimal meta-solvers via multi-objective optimization. Journal of Computational Physics, 563, 115015. https://doi.org/10.1016/j.jcp.2026.115015

Vancouver

Lee Y, Liu S, Darbon J, Karniadakis GE. Automatic discovery of optimal meta-solvers via multi-objective optimization. Journal of Computational Physics. 2026;563:115015. doi:10.1016/j.jcp.2026.115015

BibTeX
@article{Lee2026,
  author = {Lee, Youngkyu and Liu, Shanqing and Darbon, Jérôme and Karniadakis, George Em},
  title = {Automatic discovery of optimal meta-solvers via multi-objective optimization},
  journal = {Journal of Computational Physics},
  year = {2026},
  volume = {563},
  pages = {115015},
  doi = {10.1016/j.jcp.2026.115015}
}
Kostenlos starten →