arrow
返回

Fiber optic computing using distributed feedback

delete2024-03-02
delete3
delete
OA
AI
B
Brandon Redding *
J
Joseph B. Murray
J
Joseph D. Hart
Z
Zheyuan Zhu
S
Shuo Pang
R
Raktim Sarma
DOI:10.1038/s42005-024-01549-1delete
delete原文链接
delete分享
delete收藏
摘要

摘要

En 中文
The widespread adoption of machine learning and other matrix intensive computing algorithms has renewed interest in analog optical computing, which has the potential to perform large-scale matrix multiplications with superior energy scaling and lower latency than digital electronics. However, most optical techniques rely on spatial multiplexing, requiring a large number of modulators and detectors, and are typically restricted to performing a single kernel convolution operation per layer. Here, we introduce a fiber-optic computing architecture based on temporal multiplexing and distributed feedback that performs multiple convolutions on the input data in a single layer. Using Rayleigh backscattering in standard single mode fiber, we show that this technique can efficiently apply a series of random nonlinear projections to the input data, facilitating a variety of computing tasks. The approach enables efficient energy scaling with orders of magnitude lower power consumption than GPUs, while maintaining low latency and high data-throughput. Optical techniques adopted in optical computing rely on spatial multiplexing, requiring numerous integrated elements and restricting the architecture to perform a single kernel convolution per layer. The authors demonstrate a fiber-optic computing architecture based on temporal multiplexing that performs multiple convolutions in a single layer.
Keyword:
EXTREME LEARNING-MACHINE
ARTIFICIAL-INTELLIGENCE
RECOGNITION

期刊

Communications Physics 封面图
Communications Physics
IF:
5.8
论文数:
2.9K
被引数:
1.1W

机构

U
united states navy
学者数:
66
论文数: 40
被引数: 3.2K
United States Department of Defense 封面图
United States Department of Defense
学者数:
2.8W
论文数: 2.3W
被引数: 2.3W
United States Naval Research Laboratory 封面图
United States Naval Research Laboratory
学者数:
2.6K
论文数: 2.0K
被引数: 9.3K
学者 查看更多机构
引用论文

引用论文

High probability of disrupting a disulphide bridge mediated by an endogenous excited tryptophan residue
err2009-04-13
err0
errOAAI
errMaria Teresa Neves‐Petersen; Zygmunt Gryczynski; Joseph Lakowicz; Peter Fojan; Shona Pedersen; Evamaria Petersen; Steffen Bjørn Petersen
err分享
err收藏
Scalable optical learning operator
err2021-08-20
err92
errOAAI
errTegin, Ugur; Yildirim, Mustafa; Oguz, Ilker; Moser, Christophe; Psaltis, Demetri
err分享
err收藏
Rayleigh scattering optical frequency correlation in a single-mode optical fiber
err2001-01-15
err66
PREAI
errMermelstein, MD; Posey, R; Johnson, GA; Vohra, ST
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
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
Femtofarad optoelectronic integration demonstrating energy-saving signal conversion and nonlinear functions
err2019-04-15
err96
PREAI
errNozaki, Kengo; Matsuo, Shinji; Fujii, Takuro; Takeda, Koji; Shinya, Akihiko; Kuramochi, Eiichi; Notomi, Masaya
err分享
err收藏
High-speed RF spectral analysis using a Rayleigh backscattering speckle spectrometer
err2023-06-05
err8
errOAAI
errMurray, Matthew J.; Murray, Joseph B.; Schermer, Ross T.; Mckinney, Jason D.; Redding, Brandon
err分享
err收藏
学者 查看更多内容