Chaoran Huang, V. Sorger, M. Miscuglio, M. Al-Qadasi, Avilash Mukherjee, S. Shekhar, L. Chrostowski, L. Lampe, Mitchell Nichols, M. Fok, D. Brunner, A. Tait, T. F. D. Lima, B. Marquez, P. Prucnal, B. Shastri
{"title":"Prospects and applications of photonic neural networks","authors":"Chaoran Huang, V. Sorger, M. Miscuglio, M. Al-Qadasi, Avilash Mukherjee, S. Shekhar, L. Chrostowski, L. Lampe, Mitchell Nichols, M. Fok, D. Brunner, A. Tait, T. F. D. Lima, B. Marquez, P. Prucnal, B. Shastri","doi":"10.1080/23746149.2021.1981155","DOIUrl":null,"url":null,"abstract":"ABSTRACT Neural networks have enabled applications in artificial intelligence through machine learning, and neuromorphic computing. Software implementations of neural networks on conventional computers that have separate memory and processor (and that operate sequentially) are limited in speed and energy efficiency. Neuromorphic engineering aims to build processors in which hardware mimics neurons and synapses in the brain for distributed and parallel processing. Neuromorphic engineering enabled by photonics (optical physics) can offer sub-nanosecond latencies and high bandwidth with low energies to extend the domain of artificial intelligence and neuromorphic computing applications to machine learning acceleration, nonlinear programming, intelligent signal processing, etc. Photonic neural networks have been demonstrated on integrated platforms and free-space optics depending on the class of applications being targeted. Here, we discuss the prospects and demonstrated applications of these photonic neural networks. Graphical Abstract","PeriodicalId":7374,"journal":{"name":"Advances in Physics: X","volume":null,"pages":null},"PeriodicalIF":7.7000,"publicationDate":"2021-05-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"63","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advances in Physics: X","FirstCategoryId":"101","ListUrlMain":"https://doi.org/10.1080/23746149.2021.1981155","RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PHYSICS, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 63
Abstract
ABSTRACT Neural networks have enabled applications in artificial intelligence through machine learning, and neuromorphic computing. Software implementations of neural networks on conventional computers that have separate memory and processor (and that operate sequentially) are limited in speed and energy efficiency. Neuromorphic engineering aims to build processors in which hardware mimics neurons and synapses in the brain for distributed and parallel processing. Neuromorphic engineering enabled by photonics (optical physics) can offer sub-nanosecond latencies and high bandwidth with low energies to extend the domain of artificial intelligence and neuromorphic computing applications to machine learning acceleration, nonlinear programming, intelligent signal processing, etc. Photonic neural networks have been demonstrated on integrated platforms and free-space optics depending on the class of applications being targeted. Here, we discuss the prospects and demonstrated applications of these photonic neural networks. Graphical Abstract
期刊介绍:
Advances in Physics: X is a fully open-access journal that promotes the centrality of physics and physical measurement to modern science and technology. Advances in Physics: X aims to demonstrate the interconnectivity of physics, meaning the intellectual relationships that exist between one branch of physics and another, as well as the influence of physics across (hence the “X”) traditional boundaries into other disciplines including:
Chemistry
Materials Science
Engineering
Biology
Medicine