EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fast automated adjoints for spectral PDE solvers

Calum S Skene iD, Keaton J Burns iD

Proceedings of the National Academy of Sciences · 14.04.2026 · 123 (15)

DOI: 10.1073/pnas.2530440123 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 10.04.2026. Verfasst von Calum S Skene und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Applied Mathematics, University of Leeds, Leeds LS2 9JT, United Kingdom. Die Arbeit nennt 92 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Simons Foundation (SF), EC | H2020 | PRIORITY 'Excellent science' | H2020 European Research Council (ERC).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We present an automated procedure for computing model gradients for partial differential equation (PDE) solvers built on sparse spectral methods, a broad class of numerical techniques widely used in the study of fluid dynamics, continuum mechanics, waves, and pattern formation across disciplines. Our approach applies reverse-mode automatic differentiation to symbolic graph representations of PDEs and efficiently constructs adjoint solvers that retain the speed and flexibility of modern Fourier and polynomial spectral methods. We demonstrate the advantages of this approach with a comprehensive implementation in the open-source Dedalus framework. This work uniquely provides a differentiable spectral solver that supports a broad class of equations, geometries, and boundary conditions, and runs efficiently in parallel. It enables users to compute gradients and perform PDE-based optimization for a wide range of time-dependent and nonlinear models with minimal additional code. We demonstrate this system’s capabilities using canonical problems from the literature, showing both strong performance and practical utility for a wide variety of optimization tasks. By integrating automatic adjoints into a flexible solver, our work enables researchers to perform sensitivity analyses in spectral simulations with ease and efficiency.

Schlagwörter

Stichwörter der Autor:innen

PDE-based optimizationadjoint sensitivity analysisautomatic differentiationcomputational fluid dynamicsspectral methods

Autor:innen und Institutionen

  • Department of Applied Mathematics, University of Leeds, Leeds LS2 9JT, United Kingdom.
  • School of Physics and Astronomy, The University of Edinburgh, Edinburgh EH9 3FD, United Kingdom.
  • Department of Mathematics, Massachusetts Institute of Technology, Cambridge, MA 02139.
  • Center for Computational Astrophysics, Flatiron Institute, New York, NY 10010.

ORCID-Profile: Calum S Skene, Keaton J Burns

Förderung

  • Simons Foundation (SF)
  • EC | H2020 | PRIORITY 'Excellent science' | H2020 European Research Council (ERC)

Zitieren

APA

Skene, C. S., & Burns, K. J. (2026). Fast automated adjoints for spectral PDE solvers. Proceedings of the National Academy of Sciences, 123(15). https://doi.org/10.1073/pnas.2530440123

Vancouver

Skene CS, Burns KJ. Fast automated adjoints for spectral PDE solvers. Proceedings of the National Academy of Sciences. 2026;123(15). doi:10.1073/pnas.2530440123

BibTeX
@article{Skene2026,
  author = {Skene, Calum S and Burns, Keaton J},
  title = {Fast automated adjoints for spectral PDE solvers},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2026},
  volume = {123},
  number = {15},
  doi = {10.1073/pnas.2530440123}
}
Kostenlos starten →