插损鲁棒性的全复值光学神经网络  

Fully complex optical neural network with insertion-loss robustness

作　　者：陈慧彬 汤凯飞 游振宇 CHEN Hui-bin;TANG Kai-fei;YOU Zhen-yu(Institute for Photonics Technology,Quanzhou Normal University,Quanzhou 362000,China;Fujian Provincial Key Laboratory for Advanced Micro-nano Photonics Technology and Devices,Quanzhou 362000,China;College of Engineering and Applied Sciences,Nanjing University,Nanjing 210023,China)

机构地区：[1]泉州师范学院光子技术研究院,福建泉州362000 [2]福建省先进微纳光子技术与器件重点实验室,福建泉州362000 [3]南京大学现代工程与应用科学学院,江苏南京210023

出　　处：《中国光学（中英文）》2024年第4期834-841,共8页Chinese Optics

基　　金：国家自然科学基金(No.61705119)。

摘　　要：基于马赫-曾德尔干涉仪(Mach-Zehnder Interferometer,MZI)级联拓扑结构的线性光学处理器被证明是实现光学神经网络(Optical Neural Network,ONN)的重要途径,但还有不少实际问题有待解决。针对芯片制造、测试过程中可能导致的相位误差和插入损耗等问题,通过实验和理论仿真分析了几种基于MZI结构的可重构光学处理器。发现可以通过单个N×N的Clements阵列结构来实现任意酉矩阵的权重,构建稀疏连接的全复值光学神经网络,将光学深度大大降低,以实现较高的插入损耗鲁棒性。此外,对于多层光学神经网络来说,由于构建该任意酉矩阵的自由度有限,故在每一层Clements结构前面加一个相移器层,有助于将分类数据映射到更高的数据维度,能使神经网络更快速的收敛。Linear optical processors based on the cascaded topology of Mach-Zehnder Interferometer(MZI)have been demonstrated to be an important way of implementing Optical Neural Networks(ONN),but sever-al practical challenges still need resolution.Concerning issues arising from chip manufacturing and testing processes that could lead to phase errors and insertion losses,we conducted experiments and theoretical sim-ulations for various reconfigurable optical processors.We found that the weights of any arbitrary unitary mat-rix can be realized through some single N×N Clements units,that can substantially reduce the optical depth and enhance robustness against insertion losses.This approach allows for the construction of fully complex optical neural networks.Additionally,In multi-layer ONN,due to the limited degrees of freedom in con-structing this arbitrary matrix,we introduced a phase-shift layer before each layer of the Clements unit.This design aids in mapping classification data to higher-dimensional spaces,facilitating faster neural network convergence.

关 键 词：光学神经网络 MZI阵列 可重构光学处理器 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]