Lin et al. (Reports, 7 September 2018, p. 1004) reported a remarkable proposal that employs a passive, strictly linear optical setup to perform pattern classifications. But interpreting the multilayer diffractive setup as a deep neural network and advocating it as an all-optical deep learning framework are not well justified and represent a mischaracterization of the system by overlooking its defining characteristics of perfect linearity and strict passivity.
@article{arxiv.1809.08360,
title = {Comment on "All-optical machine learning using diffractive deep neural networks"},
author = {Haiqing Wei and Gang Huang and Xiuqing Wei and Yanlong Sun and Hongbin Wang},
journal= {arXiv preprint arXiv:1809.08360},
year = {2018}
}