Delocalized photonic deep learning on the internet’s edge-Reference-Cited by-同舟云学术

Delocalized photonic deep learning on the internet’s edge

Published:2022-10-21 Issue:6617 Volume:378 Page:270-276
ISSN:0036-8075
Container-title:Science
language:en
Short-container-title:Science

Author:

Sludds Alexander¹^ORCID,Bandyopadhyay Saumil¹^ORCID,Chen Zaijun¹^ORCID,Zhong Zhizhen²^ORCID,Cochrane Jared¹³^ORCID,Bernstein Liane¹^ORCID,Bunandar Darius¹^ORCID,Dixon P. Ben³^ORCID,Hamilton Scott A.³^ORCID,Streshinsky Matthew⁴,Novack Ari⁴,Baehr-Jones Tom⁴,Hochberg Michael⁴,Ghobadi Manya²,Hamerly Ryan¹⁵^ORCID,Englund Dirk¹^ORCID

Affiliation:

1. Research Laboratory of Electronics, Massachusetts Institute of Technology, Cambridge, MA 02139, USA.

2. Computer Science and Artificial Intelligence Laboratory, Massachusetts Institute of Technology, Cambridge, MA 02139, USA.

3. Lincoln Laboratory, Massachusetts Institute of Technology, Lexington, MA 02421, USA.

4. Nokia Corporation, New York, NY 10016, USA.

5. Physics and Informatics Laboratories, NTT Research Inc., Sunnyvale, CA 94085, USA.

Abstract

Advanced machine learning models are currently impossible to run on edge devices such as smart sensors and unmanned aerial vehicles owing to constraints on power, processing, and memory. We introduce an approach to machine learning inference based on delocalized analog processing across networks. In this approach, named Netcast, cloud-based “smart transceivers” stream weight data to edge devices, enabling ultraefficient photonic inference. We demonstrate image recognition at ultralow optical energy of 40 attojoules per multiply (<1 photon per multiply) at 98.8% (93%) classification accuracy. We reproduce this performance in a Boston-area field trial over 86 kilometers of deployed optical fiber, wavelength multiplexed over 3 terahertz of optical bandwidth. Netcast allows milliwatt-class edge devices with minimal memory and processing to compute at teraFLOPS rates reserved for high-power (>100 watts) cloud computers.

Publisher

American Association for the Advancement of Science (AAAS)

Subject

Multidisciplinary

Reference88 articles.

1. T. B. Brown B. Mann N. Ryder M. Subbiah J. Kaplan P. Dhariwal A. Neelakantan P. Shyam G. Sastry A. Askell S. Agarwal A. Herbert-Voss G. Krueger T. Henighan R. Child A. Ramesh D. M. Ziegler J. Wu C. Winter C. Hesse M. Chen E. Sigler M. Litwin S. Gray B. Chess J. Clark C. Berner S. McCandlish A. Radford I. Sutskever D. Amodei Language models are few-shot learners. arXiv:2005.14165 [cs.CL] (2020).

2. Grandmaster level in StarCraft II using multi-agent reinforcement learning

3. ImageNet classification with deep convolutional neural networks;Krizhevsky A.;Adv. Neural Inf. Process. Syst.,2012

4. J. Deng W. Dong R. Socher L. -J. Li K. Li L. Fei-Fei “ImageNet: A large-scale hierarchical image database ” 2009 IEEE Conference on Computer Vision and Pattern Recognition (IEEE 2009) pp. 248–255.

5. Efficient Processing of Deep Neural Networks: A Tutorial and Survey

Cited by 71 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. Surface Lattice Resonance Lasers with Epitaxial InP Gain Medium;ACS Photonics;2024-09-09

2. Silicon integrated photonic-electronic neuron for noise-resilient deep learning;Optics Express;2024-09-09

3. Intelligent Photonics: A Disruptive Technology to Shape the Present and Redefine the Future;Engineering;2024-09

4. Tunable optical matrix convolution of 20-Gbit/s QPSK 2-D data with a kernel using optical wave mixing;Optics Letters;2024-08-26

5. Photonic Crystal Cavity IQ Modulators in Thin-Film Lithium Niobate;ACS Photonics;2024-08-20