Higher-dimensional processing using a photonic tensor core with continuous-time data

New developments in hardware-based ‘accelerators’ range from electronic tensor cores and memristor-based arrays to photonic implementations. The goal of these approaches is to handle the exponentially growing computational load of machine learning, which currently requires the doubling of hardware c...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Nature photonics Ročník 17; číslo 12; s. 1080 - 1088
Hlavní autoři: Dong, Bowei, Aggarwal, Samarth, Zhou, Wen, Ali, Utku Emre, Farmakidis, Nikolaos, Lee, June Sang, He, Yuhan, Li, Xuan, Kwong, Dim-Lee, Wright, C. D, Pernice, Wolfram H. P, Bhaskaran, H
Médium: Journal Article
Jazyk:angličtina
Vydáno: London Nature Publishing Group 01.12.2023
Témata:
Accuracy
Artificial neural networks
Cardiovascular diseases
Data processing
Death
Edge computing
EKG
Electrocardiography
Frequency dependence
Frequency modulation
Hardware
Machine learning
Mathematical analysis
Multiplexing
Neural networks
Optical communication
Photonics
Radio frequency
Tensors
Wavelength
ISSN:1749-4885, 1749-4893
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:New developments in hardware-based ‘accelerators’ range from electronic tensor cores and memristor-based arrays to photonic implementations. The goal of these approaches is to handle the exponentially growing computational load of machine learning, which currently requires the doubling of hardware capability approximately every 3.5 months. One solution is increasing the data dimensionality that is processable by such hardware. Although two-dimensional data processing by multiplexing space and wavelength has been previously reported, the use of three-dimensional processing has not yet been implemented in hardware. In this paper, we introduce the radio-frequency modulation of photonic signals to increase parallelization, adding an additional dimension to the data alongside spatially distributed non-volatile memories and wavelength multiplexing. We leverage higher-dimensional processing to configure such a system to an architecture compatible with edge computing frameworks. Our system achieves a parallelism of 100, two orders higher than implementations using only the spatial and wavelength degrees of freedom. We demonstrate this by performing a synchronous convolution of 100 clinical electrocardiogram signals from patients with cardiovascular diseases, and constructing a convolutional neural network capable of identifying patients at sudden death risk with 93.5% accuracy.Radio-frequency modulation of optical signals increase the parallelization of photonic processors beyond that afforded by exploiting spatial and wavelength dimensions alone. The approach is then demonstrated on electrocardiogram signals and identifies patients at sudden death risk with 93.5% accuracy.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1749-4885
1749-4893
DOI:10.1038/s41566-023-01313-x