Abstract
An efficient nonlinear activation function for optical neural network is proposed and demonstrated by executing complex-valued algorithms and intensity measurement. Great learning capability are achieved by the inherent properties of the complex-valued optical neural network.
© 2020 The Author(s)
PDF ArticleMore Like This
Ian A. D. Williamson, Tyler W. Hughes, Momchil Minkov, Ben Bartlett, Sunil Pai, and Shanhui Fan
SM1E.2 CLEO: Science and Innovations (CLEO:S&I) 2020
Qiang Li, Shengping Liu, Yang Zhao, Wei Wang, Ye Tian, Junbo Feng, and Jin Guo
T4D.4 Asia Communications and Photonics Conference (ACP) 2020
George Mourgias-Alexandris, George Dabos, Nikolaos Passalis, Anastasios Tefas, Angelina Totovic, and Nikos Pleros
W3A.5 Optical Fiber Communication Conference (OFC) 2020