Silicon-integrated coherent neurons with 32GMAC/sec/axon compute line-rates using EAM-based input and weighting cells

Silicon-integrated coherent neurons with 32GMAC/sec/axon compute line-rates using EAM-based input and weighting cells
复制标题

具有 32GMAC/秒/轴突的硅集成相干神经元使用基于 EAM 的输入和加权单元计算线速率

DOI:
10.1109/ecoc52684.2021.9605987
复制
发表时间:
2021
期刊:
2021 European Conference on Optical Communication (ECOC)
影响因子:
--
通讯作者:
N. Pleros
N. Pleros
中科院分区:
--
文献类型:
--
作者:
G. Giamougiannis;A. Tsakyridis;G. Mourgias;M. Moralis‐Pegios;A. Totović;G. Dabos;N. Passalis;M. Kirtas;N. Bamiedakis;A. Tefas;David Lazovsky;N. Pleros

文献摘要

参考文献

被引文献

相似文献

我们通过实验证明了一个相干SiPho神经元,它依赖于EAM进行片上数据生成和加权。当该神经元作为MNIST分类器神经网络的隐藏层部署时,报告了创纪录的32GMAC/s/轴突计算率和95.91%的准确率。
We experimentally demonstrate a coherent SiPho neuron that relies on EAM for both on-chip data generation and weighting. A record-high 32GMAC/s/axon compute rate and an accuracy of 95.91% is reported, when the neuron is deployed as a hidden layer of a MNIST classifier neural network.
DOI: 10.1103/physrevapplied.11.064043
发表时间: 2018-12
影响因子: 4.6
作者:
A. Tait;T. F. D. Lima;M. Nahmias;Heidi B. Miller;Hsuan-Tung Peng;B. Shastri;P. Prucnal
通讯作者: A. Tait;T. F. D. Lima;M. Nahmias;Heidi B. Miller;Hsuan-Tung Peng;B. Shastri;P. Prucnal
DOI: 10.1038/nphoton.2017.93
发表时间: 2017-07-01
期刊: NATURE PHOTONICS
影响因子: 35
作者:
Shen, Yichen;Harris, Nicholas C.;Soljacic, Marin
通讯作者: Soljacic, Marin
DOI: 10.1109/jstqe.2018.2840448
发表时间: 2018-11-01
影响因子: 4.9
作者:
Peng, Hsuan-Tung;Nahmias, Mitchell A.;Prucnal, Paul R.
通讯作者: Prucnal, Paul R.