All-optical nonlinear activation function based on stimulated Brillouin scattering

authored by: Grigorii Slinkov, Steven Becker, Dirk Englund, Birgit Stiller
Abstract: Optical neural networks have demonstrated their potential to overcome the computational bottleneck of modern digital electronics. However, their development towards high-performing computing alternatives is hindered by one of the optical neural networks’ key components: the activation function. Most of the reported activation functions rely on opto-electronic conversion, sacrificing the unique advantages of photonics, such as resource-efficient coherent and frequency-multiplexed information encoding. Here, we experimentally demonstrate a photonic nonlinear activation function based on stimulated Brillouin scattering. It is coherent and frequency selective and can be tuned all-optically to take LEAKYRELU, SIGMOID, and QUADRATIC shape. Our design compensates for the insertion loss automatically by providing net gain as high as 20 dB, paving the way for deep optical neural networks.
Organisation(s): Institute of Photonics
External Organisation(s): Max Planck Institute for the Science of Light
Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU Erlangen-Nürnberg)
Massachusetts Institute of Technology
Type: Article
Journal: Nanophotonics
Volume: 14
Pages: 2711-2722
No. of pages: 12
ISSN: 2192-8606
Publication date: 02.08.2025
Publication status: Published
Peer reviewed: Yes
ASJC Scopus subject areas: Biotechnology, Electronic, Optical and Magnetic Materials, Atomic and Molecular Physics, and Optics, Electrical and Electronic Engineering
Electronic version(s): https://doi.org/10.1515/nanoph-2024-0513 (Access: Open)
https://doi.org/10.48550/arXiv.2401.05135 (Access: Open)
: Details in the research portal "Research@Leibniz University"

BibTeX

@article{9c5f8d704c63419baff4ef1d2ac83377,
title = "All-optical nonlinear activation function based on stimulated Brillouin scattering",
abstract = "Optical neural networks have demonstrated their potential to overcome the computational bottleneck of modern digital electronics. However, their development towards high-performing computing alternatives is hindered by one of the optical neural networks{\textquoteright} key components: the activation function. Most of the reported activation functions rely on opto-electronic conversion, sacrificing the unique advantages of photonics, such as resource-efficient coherent and frequency-multiplexed information encoding. Here, we experimentally demonstrate a photonic nonlinear activation function based on stimulated Brillouin scattering. It is coherent and frequency selective and can be tuned all-optically to take LEAKYRELU, SIGMOID, and QUADRATIC shape. Our design compensates for the insertion loss automatically by providing net gain as high as 20 dB, paving the way for deep optical neural networks.",
keywords = "Brillouin scattering, nonlinear activation function, nonlinear optics, optical fiber, optical neural network, optoacoustics, photonic neuromorphic computing",
author = "Grigorii Slinkov and Steven Becker and Dirk Englund and Birgit Stiller",
note = "Publisher Copyright: {\textcopyright} 2025 the author(s),",
year = "2025",
month = aug,
day = "2",
doi = "10.1515/nanoph-2024-0513",
language = "English",
volume = "14",
pages = "2711--2722",
journal = "Nanophotonics",
issn = "2192-8606",
publisher = "Walter de Gruyter GmbH",
number = "16",
}

Back