Prospects and applications of photonic neural networks

Advances in Physics: X

Jiahui Wang¹  Shanhui Fan¹  Bicky A. Marquez²  Bhavin J. Shastri³  Thomas Ferreira de Lima⁴  Paul R. Prucnal⁴  Chaoran Huang⁵  Volker J. Sorger⁶  Mario Miscuglio⁶  Mitchell Nichols⁷  Mohammed Al-Qadasi⁷  Avilash Mukherjee⁷  Lutz Lampe⁷  Sudip Shekhar⁷  Lukas Chrostowski⁷  Daniel Brunner⁸  Alexander N. Tait⁹  Mable P. Fok^1,10 

[1] College of Engineering, Stanford University, Stanford, CA, US;Department of Electrical Engineering, Queen’s University, Kingston, Canad;Department of Electrical Engineering, Queen’s University, Kingston, Canad;The Vector Institute, Toronto, Canad;Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, US;Department of Electrical and Computer Engineering, Princeton University, Princeton, NJ, US;Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong SAR, Chin;Department of Electrical and Computer Engineering, The George Washington University, Washington DC, DC, US;Department of Electrical and Computer Engineering, The University of British Columbia, Vancouver, Canad;Department of Physics, Engineering Physics & Astronomy, Institut FEMTO-ST, Université Bourgogne-Franche-Comté CNRS UMR, Besancon, Franc;National Institute of Standards and Technology, Boulder, Colorado, US;University of Georgia, Athens, GA, US;

关键词: Photonic neural networks;    neuromorphic photonics;    silicon photonics;    neuromorphic computing;    machine learning;    artificial intelligence;   

DOI  :  10.1080/23746149.2021.1981155

来源: Taylor & Francis

【 摘 要 】

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.

【 授权许可】

CC BY