Chinese Journal of Lasers, Volume. 47, Issue 5, 0500004(2020)

Advances and Challenges of Optical Neural Networks

Hongwei Chen1,2、*, Zhenming Yu3, Tian Zhang3, Yubin Zang1,2, Yihang Dan3, and Kun Xu3
Author Affiliations
  • 1Department of Electronic Engineering, Tsinghua University, Beijing 100084, China
  • 2Beijing National Research Center for Information Science and Technology (BNRist), Beijing 100084, China
  • 3State Key Laboratory of Information Photonics and Optical Communications,Beijing University of Posts and Telecommunications, Beijing 100876, China
  • show less
    Figures & Tables(15)
    Structure of integrated optical interference unit (OIU) and confusion matrices of vowel recognition[18].(a) Structure of OIU; (b) confusion matrices of vowel recognition by ONN and 64-bit computer
    Structure of delay line before OIU and logic and timing chart of the optical convolutional neural network[19]. (a) Structure of delay line; (b) logic and timing chart
    Structures of recurrent neural networks[20]. (a) Parallel structure; (b) serial structure
    Structure of optoelectronic recurrent neural network and classifying result of rectangular and sinusoid signals[25]. (a) Structure of optoelectronic recurrent neural network; (b) classifying result
    Graphene excitable fiber laser[29]
    Integrated photonic neuron and excitable response[30]. (a) Integrated photonic neuron; (b) excitable response
    In situ training algorithm of integrated photonic neural networks[38]
    Training results of optical neural network by adopting different algorithms[39]. (a) GA algorithm; (b) PSO algorithm
    Structure of optoelectronic nonlinear module and implementing different nonlinear functions by tunning phase of MZI[45]. (a) Structure of nonlinear module; (b) implementing different nonlinear functions
    Normalized transmission curve of optical controlled PCM[46]
    Structure of D2NN and resolution improvement of imaging[53]. (a) Structure D2NN; (b) resolution improvement result
    Structure of TS-NN[54]
    Confusion matrices of different neural network structures[54]. (a) Conventional neural network structure;(b) TS-NN without noise; (c) TS-NN with noise
    Structures of optoelectronic binarized neural network and weights mapping[55]. (a) Structure of the optoelectronic binarized neural network; (b) mapping structure of binarized weights in single polarization system; (c) mapping structure of binarized weights in polarization multiplexing system
    OSNR-BER curves of single polarization and polarization multiplexing QPSK communication systems implemented by different methods[55]. (a) Single polarization QPSK communication system; (b) polarization multiplexing QPSK communication system
    Tools

    Get Citation

    Copy Citation Text

    Hongwei Chen, Zhenming Yu, Tian Zhang, Yubin Zang, Yihang Dan, Kun Xu. Advances and Challenges of Optical Neural Networks[J]. Chinese Journal of Lasers, 2020, 47(5): 0500004

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: reviews

    Received: Nov. 26, 2019

    Accepted: Dec. 24, 2019

    Published Online: May. 12, 2020

    The Author Email: Chen Hongwei (chenhw@tsinghua.edu.cn)

    DOI:10.3788/CJL202047.0500004

    Topics