An all-optical neuron with sigmoid activation function

被引:172
作者
Mourgias-Alexandris, G. [1 ,2 ]
Tsakyridis, A. [1 ,2 ]
Passalis, N. [1 ]
Tefas, A. [1 ]
Vyrsokinos, K. [2 ,3 ]
Pleros, N. [1 ,2 ]
机构
[1] Aristotle Univ Thessaloniki, Dept Informat, Thessaloniki 54124, Greece
[2] Balkan Ctr, Ctr Interdisciplinary Res & Innovat CIRI AUTH, Bldg A&B,10th Km Thessaloniki Therrni Rd,POB 8318, GR-57001 Thessaloniki, Greece
[3] Aristotle Univ Thessaloniki, Dept Phys, Thessaloniki 54124, Greece
基金
欧盟地平线“2020”;
关键词
INTERFEROMETER;
D O I
10.1364/OE.27.009620
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
We present an all-optical neuron that utilizes a logistic sigmoid activation function, using a Wavelength-Division Multiplexing (WDM) input & weighting scheme. The activation function is realized by means of a deeply-saturated differentially-biased Semiconductor Optical Amplifier-Mach-Zehnder Interferometer (SOA-MZI) followed by a SOA-Cross-Gain-Modulation (XGM) gate. Its transfer function is both experimentally and theoretically analyzed, showing excellent agreement between theory and experiment and an almost perfect fitting with a logistic sigmoid function. The optical sigmoid transfer function is then exploited in the experimental demonstration of a photonic neuron, demonstrating successful thresholding over a 100psec-long pulse sequence with 4 different weighted-and-summed power levels. (C) 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
引用
收藏
页码:9620 / 9630
页数:11
相关论文
共 32 条
[31]  
Vaswani A., 2017, Adv. Neural Inf. Process. Syst, P30
[32]  
Xu K, 2015, PR MACH LEARN RES, V37, P2048