Photonic Perceptron Based on a Kerr Microcomb for High-Speed, Scalable, Optical Neural Networks
文献类型:期刊论文
作者 | Xu, Xingyuan1,8; Tan, Mengxi8; Corcoran, Bill7; Wu, Jiayang8; Nguyen, Thach G.6; Boes, Andreas6; Chu, Sai T.5; Little, Brent E.4; Morandotti, Roberto3; Mitchell, Arnan6 |
刊名 | LASER & PHOTONICS REVIEWS
![]() |
关键词 | Kerr micro-comb machine learning optical neural networks photonic perceptron |
ISSN号 | 1863-8880;1863-8899 |
DOI | 10.1002/lpor.202000070 |
产权排序 | 5 |
英文摘要 | Optical artificial neural networks (ONNs)-analog computing hardware tailored for machine learning-have significant potential for achieving ultra-high computing speed and energy efficiency. A new approach to architectures for ONNs based on integrated Kerr microcomb sources that is programmable, highly scalable, and capable of reaching ultra-high speeds is proposed here. The building block of the ONN-a single neuron perceptron-is experimentally demonstrated that reaches a high single-unit throughput speed of 11.9 Giga-FLOPS at 8 bits per FLOP, corresponding to 95.2 Gbps, achieved by mapping synapses onto 49 wavelengths of a microcomb. The perceptron is tested on simple standard benchmark datasets-handwritten-digit recognition and cancer-cell detection-achieving over 90% and 85% accuracy, respectively. This performance is a direct result of the record low wavelength spacing (49 GHz) for a coherent integrated microcomb source, which results in an unprecedented number of wavelengths for neuromorphic optics. Finally, an approach to scaling the perceptron to a deep learning network is proposed using the same single microcomb device and standard off-the-shelf telecommunications technology, for high-throughput operation involving full matrix multiplication for applications such as real-time massive data processing for unmanned vehicles and aircraft tracking. |
语种 | 英语 |
WOS记录号 | WOS:000555888300001 |
出版者 | WILEY-V C H VERLAG GMBH |
源URL | [http://ir.opt.ac.cn/handle/181661/93635] ![]() |
专题 | 西安光学精密机械研究所_瞬态光学技术国家重点实验室 |
通讯作者 | Moss, David J. |
作者单位 | 1.Monash Univ, Electrophoton Lab, Dept Elect & Comp Syst Engn, Clayton, Vic 3800, Australia 2.Walter & Eliza Hall Inst Med Res, Bioinformat Div, Parkville, Vic 3052, Australia 3.INRS Energie Mat & Telecommun, 1650 Blvd Lionel Boulet, Varennes, PQ J3X 1S2, Canada 4.Chinese Acad Sci, Xian Inst Opt & Precis Mech Precis Mech, Xian 710119, Peoples R China 5.City Univ Hong Kong, Dept Phys, Tat Chee Ave, Hong Kong 999077, Peoples R China 6.RMIT Univ, Sch Engn, Integrated Photon & Applicat Ctr InPAC, Melbourne, Vic 3001, Australia 7.Monash Univ, Dept Elect & Comp Syst Engn, Clayton, Vic 3800, Australia 8.Swinburne Univ Technol, Opt Sci Ctr, Hawthorn, Vic 3122, Australia |
推荐引用方式 GB/T 7714 | Xu, Xingyuan,Tan, Mengxi,Corcoran, Bill,et al. Photonic Perceptron Based on a Kerr Microcomb for High-Speed, Scalable, Optical Neural Networks[J]. LASER & PHOTONICS REVIEWS. |
APA | Xu, Xingyuan.,Tan, Mengxi.,Corcoran, Bill.,Wu, Jiayang.,Nguyen, Thach G..,...&Moss, David J.. |
MLA | Xu, Xingyuan,et al."Photonic Perceptron Based on a Kerr Microcomb for High-Speed, Scalable, Optical Neural Networks".LASER & PHOTONICS REVIEWS |
入库方式: OAI收割
来源:西安光学精密机械研究所
其他版本
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。