EUVIMEDEUVIMED entdecken
Crossref

Learning Dynamically Inspired Bases for Koopman and Transfer Operator Approximation

Gary Froyland, Kevin Kühl

SIAM Journal on Scientific Computing · 31.08.2026 · 48 (4) · C967-C990

DOI: 10.1137/25m178362x ↗

Kurz gefasst

Fachartikel in der Zeitschrift SIAM Journal on Scientific Computing, online erschienen am 30.07.2026. Verfasst von Gary Froyland und 1 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Gefördert von: Australian Research Council Laureate Fellowship, University of New South Wales, University International Postgraduate Award, University of New South Wales School of Mathematics and Statistics.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transfer and Koopman operator methods offer a framework for representing complex, nonlinear dynamical systems via linear transformations, enabling a deeper understanding of the underlying dynamics. The spectra of these operators provide important insights into system predictability and emergent behavior, although efficiently estimating them from data can be challenging. We approach this issue through the lens of general operator and representational learning, in which we approximate these linear operators using efficient finite-dimensional representations. Specifically, we machine-learn orthonormal basis functions that are dynamically tailored to the system. This learned basis provides a particularly accurate approximation of the operator’s action and enables efficient recovery of eigenfunctions and invariant measures. We illustrate our approach with examples that showcase the retrieval of spectral properties from the estimated operator, and emphasize the dynamically adaptive quality of the machine-learned basis. Reproducibility of computational results. This paper has been awarded the “SIAM Reproducibility Badge: Code and data available” as a recognition that the authors have followed reproducibility principles valued by SISC and the scientific computing community. Code and data that allow readers to reproduce the results in this paper are available at https://github.com/kevinkuhl/SABON and in the supplementary material ( SABON-main.zip [50KB]). [Formula: see text]

Förderung

  • Australian Research Council Laureate Fellowship
  • University of New South Wales, University International Postgraduate Award
  • University of New South Wales School of Mathematics and Statistics

Zitieren

APA

Froyland, G., & Kühl, K. (2026). Learning Dynamically Inspired Bases for Koopman and Transfer Operator Approximation. SIAM Journal on Scientific Computing, 48(4), C967-C990. https://doi.org/10.1137/25m178362x

Vancouver

Froyland G, Kühl K. Learning Dynamically Inspired Bases for Koopman and Transfer Operator Approximation. SIAM Journal on Scientific Computing. 2026;48(4):C967-C990. doi:10.1137/25m178362x

BibTeX
@article{Froyland2026,
  author = {Froyland, Gary and Kühl, Kevin},
  title = {Learning Dynamically Inspired Bases for Koopman and Transfer Operator Approximation},
  journal = {SIAM Journal on Scientific Computing},
  year = {2026},
  volume = {48},
  number = {4},
  pages = {C967--C990},
  doi = {10.1137/25m178362x}
}
Kostenlos starten →