Abstract
A single layer analog optical neural network based on a 44 reconfigurable optical processor is analyzed through simulation. Using a stochastic optimization algorithm and a synthetic dataset, the network yields a final accuracy of 95.6%.
© 2019 The Author(s)
PDF ArticleMore Like This
Marcus Cohen
PD4 Optical Computing (IP) 1985
Yurui Qu, Ming Zhou, Erfan Khoram, Nanfang Yu, and Zongfu Yu
FW4C.5 CLEO: Fundamental Science (CLEO:FS) 2023
H. John Caulfield
THJ7 OSA Annual Meeting (FIO) 1988