Adaptive sigmoid-like and PReLU activation functions for all-optical perceptron

被引:57
|
作者
Crnjanski, Jasna [1 ]
Krstic, Marko [1 ]
Totovic, Angelina [2 ]
Pleros, Nikos [2 ]
Gvozdic, Dejan [1 ]
机构
[1] Univ Belgrade, Sch Elect Engn, Belgrade, Serbia
[2] Aristotle Univ Thessaloniki, Ctr Interdisciplinary Res & Innovat CIRI, Dept Informat, Wireless & Photon Syst & Networks Res Grp WinPhos, TK-57001 Thessaloniki, Greece
关键词
D O I
10.1364/OL.422930
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
We present an approach for the generation of an adaptive sigmoid-like and PReLU nonlinear activation function of an all-optical perceptron, exploiting the bistability of an injection-locked Fabry-Perot semiconductor laser. The profile of the activation function can be tailored by adjusting the injection-locked side-mode order, frequency detuning of the input optical signal, Henry factor, or bias current. The universal fitting function for both families of the activation functions is presented. (C) 2021 Optical Society of America
引用
收藏
页码:2003 / 2006
页数:4
相关论文
共 50 条
  • [1] Sigmoid-like functions and root finding methods
    Petkovic, Miodrag S.
    Yun, Beong In
    APPLIED MATHEMATICS AND COMPUTATION, 2008, 204 (02) : 784 - 793
  • [2] An all-optical neuron with sigmoid activation function
    Mourgias-Alexandris, G.
    Tsakyridis, A.
    Passalis, N.
    Tefas, A.
    Vyrsokinos, K.
    Pleros, N.
    OPTICS EXPRESS, 2019, 27 (07): : 9620 - 9630
  • [3] Adaptive all-optical sigmoid activation functions for Photonic Neural Networks using Fabry-Perot laser diodes under optical injection
    Atanasijevic, Petar
    Pappas, Christos
    Banovic, Mladen
    Crnjanski, Jasna
    Tsakyridis, Apostolos
    Moralis-Pegios, Miltiadis
    Vyrsokinos, Konstantinos
    Krstic, Marko
    Mihailovic, Peda
    Petricevic, Slobodan
    Pleros, Nikos
    Gvozdic, Dejan
    2024 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXHIBITION, OFC, 2024,
  • [4] All-optical recurrent neural network with sigmoid activation function
    Mourgias-Alexandris, George
    Dabos, George
    Passalis, Nikolaos
    Tefas, Anastasios
    Totovic, Angelina
    Pleros, Nikos
    2020 OPTICAL FIBER COMMUNICATIONS CONFERENCE AND EXPOSITION (OFC), 2020,
  • [5] Reconfigurable all-optical nonlinear activation functions
    Jha, Aashu
    Huang, Chaoran
    Prucnal, Paul R.
    2020 IEEE PHOTONICS CONFERENCE (IPC), 2020,
  • [6] A coherent perceptron for all-optical learning
    Tezak, Nikolas
    Mabuchi, Hideo
    EPJ QUANTUM TECHNOLOGY, 2015, 2
  • [7] A coherent perceptron for all-optical learning
    Nikolas Tezak
    Hideo Mabuchi
    EPJ Quantum Technology, 2
  • [8] An all-optical perceptron for binary classification
    Teofilovic, Isidora
    Crnjanski, Jasna
    Banovic, Mladen
    Krstic, Marko
    Gvozdic, Dejan
    2021 29TH TELECOMMUNICATIONS FORUM (TELFOR), 2021,
  • [9] SIMPLE SIGMOID-LIKE ACTIVATION FUNCTION SUITABLE FOR DIGITAL HARDWARE IMPLEMENTATION
    KWAN, HK
    ELECTRONICS LETTERS, 1992, 28 (15) : 1379 - 1380
  • [10] Multilayer Perceptron Network with Modified Sigmoid Activation Functions
    Ebert, Tobias
    Baenfer, Oliver
    Nelles, Oliver
    ARTIFICIAL INTELLIGENCE AND COMPUTATIONAL INTELLIGENCE, PT I, 2010, 6319 : 414 - 421