Dual Optical Frequency Comb Neuron: Co‐Developing Hardware and Algorithm

Previous studies on photonic neural network have demonstrated that algorithm can inspire hardware design. This study seeks to demonstrate that hardware can also inspire algorithm design. To further exploit the advantages of photonic analog computing, the authors develop hardware and algorithm simult...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Advanced intelligent systems Ročník 5; číslo 6
Hlavní autori: Zhang, Jun, Tao, Zilong, Yan, Qiuquan, Du, Shiyin, Tang, Yuhua, Liu, Hengzhu, Wei, Ke, Zhou, Tong, Jiang, Tian
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Weinheim John Wiley & Sons, Inc 01.06.2023
Wiley
Predmet:
Ablation
Algorithms
Artificial intelligence
Artificial neural networks
Classification
Data processing
Fiber optic networks
Hardware
hardware and algorithm co-development
Inverse design
Neural networks
Optical communication
Optical frequency
optical frequency combs
photonic convolutional neural networks
Photonics
Signal processing
ISSN:2640-4567, 2640-4567
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Previous studies on photonic neural network have demonstrated that algorithm can inspire hardware design. This study seeks to demonstrate that hardware can also inspire algorithm design. To further exploit the advantages of photonic analog computing, the authors develop hardware and algorithm simultaneously for photonic convolutional neural networks. Specifically, this work developed an architecture called dual optical frequency comb neuron (DOFCN) enabled by an integrated microcomb to perform cosinusoidal nonlinear activation and vector convolution without temporal or spatial dispersion and large‐scale modulator arrays. Furthermore, DOFCN‐based composite vector convolutional neural networks (CVCNNs), an optical‐electric hybrid model, are proposed to perform classification and regression tests in signal modulation format identification and optical structure inverse design tasks, respectively. The ablation experiments show that under 4‐bit precision limit, the element‐wise activation CVCNN has 14% higher classification accuracy, 76% lower regression residuals, and 100% higher training efficiency than that of the 32‐bit standard convolutional neural network (CNN). DOFCN exhibits impressive spectral information processing ability to facilitate signal‐processing tasks related to optics and electromagnetics. Herein, a photonic neural network called the dual optical frequency comb neuron (DOFCN) is proposed, which can simultaneously perform vector convolution and nonlinear activation. Building on the DOFCN, composite vector convolutional neural networks (CVCNNs) are developed. Remarkably, the 4‐bit CVCNNs achieve better performance than 32‐bit CNNs, with a 14% higher classification accuracy and 76% lower regression residuals.
Bibliografia:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2640-4567
2640-4567
DOI:10.1002/aisy.202200417