logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
delete
arrow
返回

Universal photonic artificial intelligence acceleration

delete2025-04-09
delete2
PRE
AI
A
Ahmed, Sufi R.
R
Reza Baghdadi
M
Mikhail Bernadskiy
N
Nate Bowman
R
Ryan Braid
J
Jim Carr
C
Chen Chen
P
Pietro Ciccarella
M
Matthew O. T. Cole
J
John P. Cooke
D
Desai, Kishor
C
Carlos Dorta
J
Jonathan Elmhurst
B
Bryce Gardiner
E
Elliot Greenwald
S
Shashank Gupta
P
Parry Husbands
B
Brian Jones
A
Anthony Kopa
H
Ho John Lee
A
Arulselvan Madhavan
A
Adam Mendrela
N
Nicholas Moore
L
Lakshmi Nair
A
Aditya Om
P
Patel, Subie
R
Rutayan Patro
R
Rob Pellowski
E
Esha Radhakrishnani
S
Sandeep Sane
N
Nicholas Sarkis
J
Joe Stadolnik
M
Mykhailo Tymchenko
G
Gongyu Wang
K
Kurt Winikka
A
Alexandra Wleklinski
J
Josh Zelman
R
R. Ho
R
Ritesh Jain
A
Ayon Basumallik
D
Darius Bunandar
N
Nicholas C. Harris *
DOI：10.1038/s41586-025-08854-xdelete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
过去十年，光子学研究已探索加速张量运算，该运算为人工智能（AI）和深度学习1, 2, 3-4奠定基础，并将其视为提升能源效率与性能5, 6, 7, 8, 9, 10, 11, 12, 13-14的途径。该领域的主要动机在于寻找替代技术，以在超越摩尔定律和 Dennard 缩放的时代延续计算进展15, 16, 17, 18-19。尽管取得了这些进展，但尚未有光子芯片达到实践人工智能应用所需的精度，且演示仅限于简化的基准任务。本文介绍一种光子人工智能处理器，其可执行包括 ResNet3 和 BERT20,21在内的高级人工智能模型，以及由 DeepMind 原创演示的 Atari 深度强化学习算法22。该处理器在众多工作负载中实现了接近电子设备的精度，标志着光子计算在竞争中与成熟电子人工智能加速器23的显著突破，并是实现后晶体管计算技术发展的关键一步。
Keyword:
DEEP
LEVEL

期刊

Nature 封面图
Nature
IF:
48.5
论文数:
1.8W
被引数:
96.5W

机构

L
lightmatter
学者数:
46
论文数: 2
被引数: 3
O
openai
学者数:
53
论文数: 27
被引数: 4
引用论文

引用论文

A 9-bit, 45 mW, 0.05 mm2 Source-Series-Terminated DAC Driver With Echo Canceller in 22-nm CMOS for In-Vehicle Communication一款具有回声消除功能的9位、功耗45 mW、面积0.05 mm2的源端串联终止型DAC驱动器，采用22-nm CMOS工艺，用于车载通信。
errIEEE Solid-State Circuits Letters
IF0
err2021-01-01
err0
PREAI
errHossein Ghafarian; Suhas Shivapakash; Sanaz Mortazavi; Philipp Scholz; Nima Lotfi; Friedel Gerfers
err分享
err收藏
A 10.1-ENOB, 6.2-fJ/conv.-step, 500-MS/s, Ringamp-Based Pipelined-SAR ADC With Background Calibration and Dynamic Reference Regulation in 16-nm CMOS
errIEEE Journal of Solid-State Circuits
IF0
err2022-04-01
err0
PREAI
errJorge Lagos; Nereo Markulic; Benjamin Hershberg; Davide Dermit; Mithlesh Shrivas; Ewout Martens; Jan Craninckx
err分享
err收藏
All-optical machine learning using diffractive deep neural networks基于衍射深度神经网络的全光机器学习
errScience
IF0
err2018-09-07
err0
errOAAI
errXing Lin; Yair Rivenson; Nezih T. Yardimci; Muhammed Veli; Yi Luo; Mona Jarrahi; Aydogan Ozcan
err分享
err收藏
Photonics for artificial intelligence and neuromorphic computing用于人工智能和神经形态计算的光子学
errNATURE PHOTONICS
IF32.9
err2021-01-29
err995
errOAAI
errShastri, Bhavin J.; Tait, Alexander N.; de Lima, T. Ferreira; Pernice, Wolfram H. P.; Bhaskaran, Harish; Wright, C. D.; Prucnal, Paul R.
err分享
err收藏
Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference用于有效的仅整数算术推理的神经网络的量化和训练
err2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
IF0
err2018-06-01
err0
errOAAI
errBenoit Jacob; Skirmantas Kligys; Bo Chen; Menglong Zhu; Matthew Tang; Andrew Howard; Hartwig Adam; Dmitry Kalenichenko
err分享
err收藏
Cats and dogs猫和狗
err2012 IEEE Conference on Computer Vision and Pattern Recognition
IF0
err2012-06-01
err0
PREAI
errO. M. Parkhi; A. Vedaldi; A. Zisserman; C. V. Jawahar
err分享
err收藏
Integrated non-reciprocal magneto-optics with ultra-high endurance for photonic in-memory computing
errNATURE PHOTONICS
IF32.9
err2024-10-23
err0
errOAAI
errPintus, Paolo; Dumont, Mario; Shah, Vivswan; Murai, Toshiya; Shoji, Yuya; Huang, Duanni; Moody, Galan; Bowers, John E.; Youngblood, Nathan
err分享
err收藏
ImageNet Classification with Deep Convolutional Neural Networks基于深度卷积神经网络的ImageNet分类
errCOMMUNICATIONS OF THE ACM
IF12.2
err2017-05-24
err8.3W
errOAAI
errKrizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E.
err分享
err收藏
Deep learning with coherent VCSEL neural networks基于相干VCSEL神经网络的深度学习
errNATURE PHOTONICS
IF32.9
err2023-07-17
err49
PREAI
errChen, Zaijun; Sludds, Alexander; Davis III, Ronald; Christen, Ian; Bernstein, Liane; Ateshian, Lamia; Heuser, Tobias; Heermeier, Niels; Lott, James A.; Reitzenstein, Stephan; Hamerly, Ryan; Englund, Dirk
err分享
err收藏
The future of computing beyond Moore's Law
errPHILOSOPHICAL TRANSACTIONS OF THE ROYAL SOCIETY A-MATHEMATICAL PHYSICAL AND ENGINEERING SCIENCES
IF3.7
err2020-01-20
err331
errOAAI
errShalf, John
err分享
err收藏
学者 查看更多内容