Training of Photonic Neural Networks through In Situ Backpropagation

被引：0
作者：
Hughes, Tyler W. [1 ]
Minkov, Momchil [1 ]
Williamson, Ian A. D. [1 ]
Shi, Yu [1 ]
Fan, Shanhui [1 ]
机构：
[1] Stanford Univ, Ginzton Lab, Stanford, CA 94305 USA
来源：
2019 CONFERENCE ON LASERS AND ELECTRO-OPTICS (CLEO) | 2019年
关键词：
D O I：
暂无
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：
We provide a protocol for training photonic neural networks based on adjoint methods. The gradient of the network with respect to its tunable degrees of freedom is computed by physically backpropagating an optical error signal. (C) 2019 The Author(s)
引用
收藏
页数：2
相关论文
共 50 条
  • [41] MULTIPLE TRAINING CONCEPT FOR BACKPROPAGATION NEURAL NETWORKS FOR USE IN ASSOCIATIVE MEMORIES
    WANG, YF
    CRUZ, JB
    MULLIGAN, JH
    NEURAL NETWORKS, 1993, 6 (08) : 1169 - 1175
  • [42] A Modified Gradient-based Backpropagation Training Method for Neural Networks
    Mu, Xuewen
    Zhang, Yaling
    2009 IEEE INTERNATIONAL CONFERENCE ON GRANULAR COMPUTING ( GRC 2009), 2009, : 450 - +
  • [43] Generalized backpropagation algorithm for training second-order neural networks
    Fan, Fenglei
    Cong, Wenxiang
    Wang, Ge
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN BIOMEDICAL ENGINEERING, 2018, 34 (05)
  • [44] An Efficient non-Backpropagation Method for Training Spiking Neural Networks
    Guo, Shiqi
    Lin, Tong
    2021 IEEE 33RD INTERNATIONAL CONFERENCE ON TOOLS WITH ARTIFICIAL INTELLIGENCE (ICTAI 2021), 2021, : 192 - 199
  • [45] Navigating beyond backpropagation: on alternative training methods for deep neural networks
    Birjais, Roshan
    Wang, Kevin I-Kai
    Abdulla, Waleed
    KNOWLEDGE AND INFORMATION SYSTEMS, 2025,
  • [46] Dual adaptive training of photonic neural networks
    Zheng, Ziyang
    Duan, Zhengyang
    Chen, Hang
    Yang, Rui
    Gao, Sheng
    Zhang, Haiou
    Xiong, Hongkai
    Lin, Xing
    NATURE MACHINE INTELLIGENCE, 2023, 5 (10) : 1119 - +
  • [47] Active training of backpropagation neural networks using the learning by experimentation methodology
    Fu-Ren Lin
    Michael J. Shaw
    Annals of Operations Research, 1997, 75 : 105 - 122
  • [48] A MODIFIED BACKPROPAGATION ALGORITHM FOR TRAINING NEURAL NETWORKS ON DATA WITH ERROR BARS
    GERNOTH, KA
    CLARK, JW
    COMPUTER PHYSICS COMMUNICATIONS, 1995, 88 (01) : 1 - 22
  • [49] Evolutionary algorithms as an alternative to backpropagation for supervised training of Biophysical Neural Networks and Neural ODEs
    Hazelden, James
    Liu, Yuhan Helena
    Shlizerman, Eli
    Shea-Brown, Eric
    arXiv, 2023,
  • [50] Active training of backpropagation neural networks using the learning by experimentation methodology
    Lin, FR
    Shaw, MJ
    ANNALS OF OPERATIONS RESEARCH, 1997, 75 (0) : 105 - 122
12345