EUVIMEDEUVIMED entdecken
Crossref

On-chip input-hidden-layer-degenerate optical diffractive nonlinear neural network

Wenguang Xu, Bo Wu, Shiji Zhang, Hailong Zhou, Yilun Wang, Jianji Dong, Xinliang Zhang

Optica · 20.01.2026 · 13 (1) · 172

DOI: 10.1364/optica.578963 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Optica, online erschienen am 20.01.2026. Verfasst von Wenguang Xu und 6 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 60 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Integrated photonic computing offers compelling advantages, including ultrahigh bandwidth, ultralow latency, and inherently low power consumption, positioning it as a promising platform for artificial intelligence (AI) applications. However, the input scale of current photonic computing chips remains fundamentally constrained by their one-dimensional interface. Here, we propose and experimentally demonstrate an input-hidden-layer-degenerate optical diffractive nonlinear neural network (IHD-ODN 3 ), which expands the input dimension of on-chip optical networks from N to N 2 within a compact footprint. By structurally merging the input layer with linear diffractive hidden layers, input data are directly encoded into optical neurons. Leveraging the intrinsic properties of diffraction, multilayer diffraction interactions among these neurons promote sufficient data mixing, thereby executing an equivalent nonlinear function and mapping the data into a high-dimensional space with a footprint of 0.5mm 2 . We validate the nonlinear computational capability of IHD-ODN 3 through four-class spiral and handwritten digit classification tasks, achieving accuracies of 88% and 98%, respectively. Furthermore, we demonstrate high-performance image compression and reconstruction, with a compression ratio of up to 256:1. This input-hidden-layer-degenerate architecture not only unlocks N 2 dimensional input encoding, but also delivers low-power, highly integrated and programmable optical nonlinear computing with linear photonic hardware. Our approach is fully compatible with mainstream linear optical neural network architectures, paving the way for next generation of large-scale, highly integrated nonlinear optical deep neural networks.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Xu, W., Wu, B., Zhang, S., Zhou, H., Wang, Y., Dong, J., & Zhang, X. (2026). On-chip input-hidden-layer-degenerate optical diffractive nonlinear neural network. Optica, 13(1), 172. https://doi.org/10.1364/optica.578963

Vancouver

Xu W, Wu B, Zhang S, Zhou H, Wang Y, Dong J, et al. On-chip input-hidden-layer-degenerate optical diffractive nonlinear neural network. Optica. 2026;13(1):172. doi:10.1364/optica.578963

BibTeX
@article{Xu2026,
  author = {Xu, Wenguang and Wu, Bo and Zhang, Shiji and Zhou, Hailong and Wang, Yilun and Dong, Jianji and Zhang, Xinliang},
  title = {On-chip input-hidden-layer-degenerate optical diffractive nonlinear neural network},
  journal = {Optica},
  year = {2026},
  volume = {13},
  number = {1},
  pages = {172},
  doi = {10.1364/optica.578963}
}
Kostenlos starten →