神经形态硅光子学与50 GHz平铺矩阵乘法深度学习应用

IF 20.6 1区 物理与天体物理 Q1 OPTICS Advanced Photonics Pub Date : 2023-01-01 DOI:10.1117/1.AP.5.1.016004
G. Giamougiannis, A. Tsakyridis, M. Moralis‐Pegios, G. Mourgias-Alexandris, A. Totović, G. Dabos, M. Kirtas, N. Passalis, A. Tefas, D. Kalavrouziotis, D. Syrivelis, P. Bakopoulos, E. Mentovich, David Lazovsky, N. Pleros
{"title":"神经形态硅光子学与50 GHz平铺矩阵乘法深度学习应用","authors":"G. Giamougiannis, A. Tsakyridis, M. Moralis‐Pegios, G. Mourgias-Alexandris, A. Totović, G. Dabos, M. Kirtas, N. Passalis, A. Tefas, D. Kalavrouziotis, D. Syrivelis, P. Bakopoulos, E. Mentovich, David Lazovsky, N. Pleros","doi":"10.1117/1.AP.5.1.016004","DOIUrl":null,"url":null,"abstract":"Abstract. The explosive volume growth of deep-learning (DL) applications has triggered an era in computing, with neuromorphic photonic platforms promising to merge ultra-high speed and energy efficiency credentials with the brain-inspired computing primitives. The transfer of deep neural networks (DNNs) onto silicon photonic (SiPho) architectures requires, however, an analog computing engine that can perform tiled matrix multiplication (TMM) at line rate to support DL applications with a large number of trainable parameters, similar to the approach followed by state-of-the-art electronic graphics processing units. Herein, we demonstrate an analog SiPho computing engine that relies on a coherent architecture and can perform optical TMM at the record-high speed of 50 GHz. Its potential to support DL applications, where the number of trainable parameters exceeds the available hardware dimensions, is highlighted through a photonic DNN that can reliably detect distributed denial-of-service attacks within a data center with a Cohen’s kappa score-based accuracy of 0.636.","PeriodicalId":33241,"journal":{"name":"Advanced Photonics","volume":null,"pages":null},"PeriodicalIF":20.6000,"publicationDate":"2023-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"Neuromorphic silicon photonics with 50 GHz tiled matrix multiplication for deep-learning applications\",\"authors\":\"G. Giamougiannis, A. Tsakyridis, M. Moralis‐Pegios, G. Mourgias-Alexandris, A. Totović, G. Dabos, M. Kirtas, N. Passalis, A. Tefas, D. Kalavrouziotis, D. Syrivelis, P. Bakopoulos, E. Mentovich, David Lazovsky, N. Pleros\",\"doi\":\"10.1117/1.AP.5.1.016004\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Abstract. The explosive volume growth of deep-learning (DL) applications has triggered an era in computing, with neuromorphic photonic platforms promising to merge ultra-high speed and energy efficiency credentials with the brain-inspired computing primitives. The transfer of deep neural networks (DNNs) onto silicon photonic (SiPho) architectures requires, however, an analog computing engine that can perform tiled matrix multiplication (TMM) at line rate to support DL applications with a large number of trainable parameters, similar to the approach followed by state-of-the-art electronic graphics processing units. Herein, we demonstrate an analog SiPho computing engine that relies on a coherent architecture and can perform optical TMM at the record-high speed of 50 GHz. Its potential to support DL applications, where the number of trainable parameters exceeds the available hardware dimensions, is highlighted through a photonic DNN that can reliably detect distributed denial-of-service attacks within a data center with a Cohen’s kappa score-based accuracy of 0.636.\",\"PeriodicalId\":33241,\"journal\":{\"name\":\"Advanced Photonics\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":20.6000,\"publicationDate\":\"2023-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Advanced Photonics\",\"FirstCategoryId\":\"101\",\"ListUrlMain\":\"https://doi.org/10.1117/1.AP.5.1.016004\",\"RegionNum\":1,\"RegionCategory\":\"物理与天体物理\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"OPTICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advanced Photonics","FirstCategoryId":"101","ListUrlMain":"https://doi.org/10.1117/1.AP.5.1.016004","RegionNum":1,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OPTICS","Score":null,"Total":0}
引用次数: 6

摘要

摘要深度学习(DL)应用的爆炸式增长引发了一个计算时代,神经形态光子平台有望将超高速和能效证书与受大脑启发的计算原语相结合。然而,将深度神经网络(DNN)转移到硅光子(SiPho)架构上需要一个模拟计算引擎,该引擎可以以线速率执行平铺矩阵乘法(TMM),以支持具有大量可训练参数的DL应用,类似于最先进的电子图形处理单元所遵循的方法。在此,我们展示了一个模拟SiPho计算引擎,该引擎依赖于相干架构,可以以创纪录的50 GHz高速执行光学TMM。通过光子DNN,它可以可靠地检测数据中心内的分布式拒绝服务攻击,基于Cohen’s kappa评分的准确度为0.636,突出了它支持DL应用的潜力,在DL应用中,可训练参数的数量超过了可用的硬件尺寸。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
Neuromorphic silicon photonics with 50 GHz tiled matrix multiplication for deep-learning applications
Abstract. The explosive volume growth of deep-learning (DL) applications has triggered an era in computing, with neuromorphic photonic platforms promising to merge ultra-high speed and energy efficiency credentials with the brain-inspired computing primitives. The transfer of deep neural networks (DNNs) onto silicon photonic (SiPho) architectures requires, however, an analog computing engine that can perform tiled matrix multiplication (TMM) at line rate to support DL applications with a large number of trainable parameters, similar to the approach followed by state-of-the-art electronic graphics processing units. Herein, we demonstrate an analog SiPho computing engine that relies on a coherent architecture and can perform optical TMM at the record-high speed of 50 GHz. Its potential to support DL applications, where the number of trainable parameters exceeds the available hardware dimensions, is highlighted through a photonic DNN that can reliably detect distributed denial-of-service attacks within a data center with a Cohen’s kappa score-based accuracy of 0.636.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
CiteScore
22.70
自引率
1.20%
发文量
49
审稿时长
18 weeks
期刊介绍: Advanced Photonics is a highly selective, open-access, international journal that publishes innovative research in all areas of optics and photonics, including fundamental and applied research. The journal publishes top-quality original papers, letters, and review articles, reflecting significant advances and breakthroughs in theoretical and experimental research and novel applications with considerable potential. The journal seeks high-quality, high-impact articles across the entire spectrum of optics, photonics, and related fields with specific emphasis on the following acceptance criteria: -New concepts in terms of fundamental research with great impact and significance -State-of-the-art technologies in terms of novel methods for important applications -Reviews of recent major advances and discoveries and state-of-the-art benchmarking. The journal also publishes news and commentaries highlighting scientific and technological discoveries, breakthroughs, and achievements in optics, photonics, and related fields.
期刊最新文献
Large-scale distributed diffractive-interference hybrid photonic chiplets Coherence entropy during propagation through complex media Authentication through residual attention-based processing of tampered optical responses Controlling the hidden parity in vectorial light with metasurfaces Observing the collapse of super-Bloch oscillations in strong-driving photonic temporal lattices
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1