Optoelectronic Demonstration of the Nonlinear Activation Module for Optical Neural Networks

被引：0
作者：
Zang, Yubin [1 ,2 ]
Li, Simin [3 ,4 ]
Hua, Boyu
Lin, Zhipeng
Zhang, Fangzheng
Zhang, Zuxing [1 ,2 ]
机构：
[1] Nanjing Univ Posts & Telecommun, Coll Elect & Opt Engn, Adv Photon Technol Lab, Future Technol, Nanjing 210023, Peoples R China
[2] Nanjing Univ Posts & Telecommun, Coll Flexible Elect, Future Technol, Nanjing 210023, Peoples R China
[3] Nanjing Univ Aeronaut & Astronaut, Coll Elect & Informat Engn, Key Lab Radar Imaging & Microwave Photon, Nanjing, Peoples R China
[4] Nanjing Univ Aeronaut & Astronaut, Key Lab Radar Imaging & Microwave Photon, Nanjing, Peoples R China
来源：
IEEE PHOTONICS TECHNOLOGY LETTERS | 2025年 / 37卷 / 02期
关键词：
Optical computing; Optical films; Adaptive optics; Neural networks; Optical network units; Voltage; Optical attenuators; Optical pulses; Optical imaging; Optical materials; Nonlinear activation module; optical neural networks; tunable attenuator; optical computing; EXPERIMENTAL REALIZATION;
D O I：
10.1109/LPT.2024.3517159
中图分类号：
TM [电工技术]; TN [电子技术、通信技术];
学科分类号：
0808 ; 0809 ;
摘要：
In this letter, we experimentally demonstrated the viability of the optoelectronic nonlinear activation module in optical neural networks. By adopting the beam splitter, photo-diode, electronic processing unit and tunable attenuator, nonlinear activation function ReLU which is widely used in the state of the art neural networks can be implemented in an optoelectronic way. Compared with other optical and optoelectronic implementations of nonlinear activation function, this method has advantages in expanses and complexity. This method also has potential to be applied into multiple types of optical neural networks to further increase the prediction accuracy.
引用
收藏
页码：109 / 112
页数：4
相关论文
共 17 条
  • [1] Bandyopadhyay S, 2022, Arxiv, DOI arXiv:2208.01623
  • [2] Reinforcement learning in a large-scale photonic recurrent neural network
    Bueno, J.
    Maktoobi, S.
    Froehly, L.
    Fischer, I.
    Jacquot, M.
    Larger, L.
    Brunner, D.
    [J]. OPTICA, 2018, 5 (06): : 756 - 760
  • [3] All-optical reservoir computing
    Duport, Francois
    Schneider, Bendix
    Smerieri, Anteo
    Haelterman, Marc
    Massar, Serge
    [J]. OPTICS EXPRESS, 2012, 20 (20): : 22783 - 22795
  • [4] All-optical spiking neurosynaptic networks with self-learning capabilities
    Feldmann, J.
    Youngblood, N.
    Wright, C. D.
    Bhaskaran, H.
    Pernice, W. H. P.
    [J]. NATURE, 2019, 569 (7755) : 208 - +
  • [5] Experimental realization of arbitrary activation functions for optical neural networks
    Frad, Monireh Moayedi Pour
    Williamson, Ian A. D.
    Edwards, Matthew
    Liu, Ke
    Pai, Sunil
    Bartlett, Ben
    Minkov, Momchil
    Hughes, Tyler W.
    Fan, Shanhui
    Nguyen, Thien-An
    [J]. OPTICS EXPRESS, 2020, 28 (08): : 12138 - 12148
  • [6] Graphene-assisted multiple-input high-base optical computing
    Hu, Xiao
    Wang, Andong
    Zeng, Mengqi
    Long, Yun
    Zhu, Long
    Fu, Lei
    Wang, Jian
    [J]. SCIENTIFIC REPORTS, 2016, 6
  • [7] Lim GK, 2011, NAT PHOTONICS, V5, P554, DOI [10.1038/NPHOTON.2011.177, 10.1038/nphoton.2011.177]
  • [8] Optoelectronic Reservoir Computing
    Paquot, Y.
    Duport, F.
    Smerieri, A.
    Dambre, J.
    Schrauwen, B.
    Haelterman, M.
    Massar, S.
    [J]. SCIENTIFIC REPORTS, 2012, 2
  • [9] EXPERIMENTAL REALIZATION OF ANY DISCRETE UNITARY OPERATOR
    RECK, M
    ZEILINGER, A
    BERNSTEIN, HJ
    BERTANI, P
    [J]. PHYSICAL REVIEW LETTERS, 1994, 73 (01) : 58 - 61
  • [10] Demonstration of a 4 x 4-port universal linear circuit
    Ribeiro, Antonio
    Ruocco, Alfonso
    Vanacker, Laurent
    Bogaerts, Wim
    [J]. OPTICA, 2016, 3 (12): : 1348 - 1357
12