EUVIMEDEUVIMED entdecken
Crossref

Delay-embedding for recurrence in optical reservoir computing

Anja Bartelmei, Maximilian Zier, Jörg Schumacher, Stefan Sinzinger

Optics Express · 27.07.2026 · 34 (15) · 28496

DOI: 10.1364/oe.595960 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Optics Express, online erschienen am 23.07.2026. Verfasst von Anja Bartelmei und 3 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: Carl Zeiss Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Time-delay embedding is a prominent method in neural networks to improve the performance of predictions of the dynamics of nonlinear systems. Applied to reservoir computing, a recurrent machine learning model in which the output weights are trained only, an improved prediction performance for a broader range of hyperparameters can be realized. Here, we present an optical implementation of reservoir computing on the basis of a spatial light modulator. In addition to the high parallelism of the optical setup and the fast processing capabilities with scalable numbers of neurons, a recursive cascading effect is generated for successful optical delay embedding. Our laboratory experiment is evaluated for the one-dimensional Kuramoto-Sivashinsky equation and is found to perform better than a classical implementation of reservoir computing.

Förderung

  • Carl Zeiss Foundation

Zitieren

APA

Bartelmei, A., Zier, M., Schumacher, J., & Sinzinger, S. (2026). Delay-embedding for recurrence in optical reservoir computing. Optics Express, 34(15), 28496. https://doi.org/10.1364/oe.595960

Vancouver

Bartelmei A, Zier M, Schumacher J, Sinzinger S. Delay-embedding for recurrence in optical reservoir computing. Optics Express. 2026;34(15):28496. doi:10.1364/oe.595960

BibTeX
@article{Bartelmei2026,
  author = {Bartelmei, Anja and Zier, Maximilian and Schumacher, Jörg and Sinzinger, Stefan},
  title = {Delay-embedding for recurrence in optical reservoir computing},
  journal = {Optics Express},
  year = {2026},
  volume = {34},
  number = {15},
  pages = {28496},
  doi = {10.1364/oe.595960}
}
Kostenlos starten →