Training neural networks with end-to-end optical backpropagation  

作　　者：James Spall Xianxin Guo Alexander I.Lvovsky 

机构地区：[1]University of Oxford,Clarendon Laboratory,Oxford,United Kingdom [2]Lumai Ltd.,Wood Centre for Innovation,Oxford,United Kingdom

出　　处：《Advanced Photonics》2025年第1期31-40,共10页先进光子学(英文)

基　　金：supported by the Innovate UK Smart (Grant No. 10043476);support from the Royal Commission for the Exhibition of 1851 Research Fellowship。

摘　　要：Optics is an exciting route for the next generation of computing hardware for machine learning,promising several orders of magnitude enhancement in both computational speed and energy efficiency.However, reaching the full capacity of an optical neural network(NN) necessitates that the computing be implemented optically not only for inference but also for training. The primary algorithm for network training is backpropagation, in which the calculation is performed in the order opposite to the information flow for inference. Although straightforward in a digital computer, the optical implementation of backpropagation has remained elusive, particularly because of the conflicting requirements for the optical element that implements the nonlinear activation function. We address this challenge for the first time, we believe, with a surprisingly simple scheme, employing saturable absorbers for the role of activation units. Our approach is adaptable to various analog platforms and materials and demonstrates the possibility of constructing NNs entirely reliant on analog optical processes for both training and inference tasks.

关 键 词：optical computing optical neural networks machine learning 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]