High‐Bit‐Efficiency TOPS Optical Tensor Convolutional Accelerator Using Microcombs

Abstract Tensor convolution is a fundamental operation in convolutional neural networks, especially for processing tensors, which are prevalent in real‐world applications. Current methods often convert tensor convolutions into matrix multiplications, leading to data replication, additional memory usage and increased hardware complexity. Here, a high‐bit‐efficiency optical tensor convolution accelerator with reduced data redundancy and lower memory consumption is presented. The bit‐efficiency of the optical tensor convolution accelerator is first explored, significantly improving its effective computing power by utilizing the spatial dimension. Consequently, the optical tensor convolutional accelerator operates at speeds exceeding 3 Tera Operations Per Second (TOPS)—the fastest single‐kernel optical convolutional accelerator to date, to the best of authors' knowledge. Its performance is validated on handwritten digit recognition and histopathologic cancer detection tasks, achieving 93.8% and 77% accuracy, respectively, closely matching in‐silico results. This approach simultaneously multiplexes the physical dimensions—wavelength, time, and space—and leverages the parallelism and high throughput of light, enabling efficient optical processing of tensor data with significant computational power.

Authors

Institutions

Publication Details

Journal
Laser & Photonics Review
Published
2025-02-04
DOI
https://doi.org/10.1002/lpor.202401975
Citations
14
Primary Topic
Neural Networks and Reservoir Computing
Type
article
Field-Weighted Citation Impact
34.73

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

High‐Bit‐Efficiency TOPS Optical Tensor Convolutional Accelerator Using Microcombs

Sirui Huang, Roberto Morandotti, Xingyuan Xu, Chan Xia et al.
14 citations
Laser & Photonics Review
Neural Networks and Reservoir Computing
34.73
article

High‐Bit‐Efficiency TOPS Optical Tensor Convolutional Accelerator Using Microcombs

Sirui Huang, Roberto Morandotti, Xingyuan Xu, Chan Xia, Yunping Bai, Kun Xu, Yixuan Zheng, Sai T. Chu, Shuai Wang, Xiaotian Zhu, Chaoran Huang, Shifan Chen, David Moss, Yuyang Liu, Yifu Xu, Xiaoyan Xu, Brent E. Little, Zhihui Liu
article en
14 citations

Abstract

Abstract Tensor convolution is a fundamental operation in convolutional neural networks, especially for processing tensors, which are prevalent in real‐world applications. Current methods often convert tensor convolutions into matrix multiplications, leading to data replication, additional memory usage and increased hardware complexity. Here, a high‐bit‐efficiency optical tensor convolution accelerator with reduced data redundancy and lower memory consumption is presented. The bit‐efficiency of the optical tensor convolution accelerator is first explored, significantly improving its effective computing power by utilizing the spatial dimension. Consequently, the optical tensor convolutional accelerator operates at speeds exceeding 3 Tera Operations Per Second (TOPS)—the fastest single‐kernel optical convolutional accelerator to date, to the best of authors' knowledge. Its performance is validated on handwritten digit recognition and histopathologic cancer detection tasks, achieving 93.8% and 77% accuracy, respectively, closely matching in‐silico results. This approach simultaneously multiplexes the physical dimensions—wavelength, time, and space—and leverages the parallelism and high throughput of light, enabling efficient optical processing of tensor data with significant computational power.

Laser & Photonics ReviewVol. 19(9)
Beijing University of Posts and Telecommunications (CN), City University of Hong Kong (HK), Chinese University of Hong Kong (HK), Institut National de la Recherche Scientifique (CA), China Telecom (China) (CN), China Telecom (CN), Swinburne University of Technology (AU)
National Natural Science Foundation of China, Australian Research Council, National Key Research and Development Program of China
Openalex Percentile: Top 0%
Neural Networks and Reservoir Computing
34.73
DOISourceOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.