Advances and Challenges of Optical Neural Networks
Advances and Challenges of Optical Neural Networks
复制标题
DOI：
10.3788/cjl202047.0500004
复制
发表时间：
2020-05-01
影响因子：
1.7
通讯作者：
Xu Kun
中科院分区：
文献类型：
--
作者：
Chen Hongwei;Yu Zhenming;Xu Kun
神经网络作为人工智能领域最具代表性的技术之一，正朝着高计算速度、低功耗的方向快速发展。由于电子设备本身的局限性，电子实现的神经网络很难进一步提高这两个性能。光学神经网络可以将光电子技术和神经网络模型联合收割机结合起来，为突破这一瓶颈提供途径。为了更好地了解光学神经网络的发展历史、前沿和未来，本文对前馈型、递归型和脉冲型光学神经网络进行了综述。揭示了光学神经网络在原位训练、非线性计算、扩展规模和应用等方面面临的挑战和发展趋势。
Neural networks, as one of the most representative techniques in artificial intelligence, have been in rapid development towards high computational speed and low power cost. Due to intrinsic limitations brought by electronic devices, it can be hard for electronic implemented neural networks to further improve these two performances. Optical neural networks can combine both optoelectronic technique and neural network model to provide ways to break the bottleneck. In order to have a brighter view on the history, frontiers and future of optical neural networks, optical neural networks of feed -forward, recurrent and spiking models arc illustrated in this paper. Challenges and future trends of optical neural networks on in situ training, nonlinear computing, expanding scale and applications will thus be revealed.