120 GOPS Photonic tensor core in thin-film lithium niobate for inference and in situ training

Zhongjin Lin, Bhavin J. Shastri, Shangxuan Yu, Jingxiang Song, Yuntao Zhu, Arman Safarnejadian, Wangning Cai, Yanmei Lin, Wei Ke, Mustafa Hammood, Tianye Wang, Mengyue Xu, Zibo Zheng, Mohammed Al-Qadasi, Omid Esmaeeli, Mohamed Rahim, Grzegorz Pakulski, Jens Schmid, Pedro Barrios, Weihong Jiang, Hugh Morison, Matthew Mitchell, Xun Guan, Nicolas A. F. Jaeger, Leslie A. Rusch, Sudip Shekhar, Wei Shi, Siyuan Yu, Xinlun Cai () and Lukas Chrostowski ()
Additional contact information
Zhongjin Lin: The University of British Columbia
Bhavin J. Shastri: Queen’s University
Shangxuan Yu: The University of British Columbia
Jingxiang Song: The University of British Columbia
Yuntao Zhu: Sun Yat-sen University
Arman Safarnejadian: Université Laval
Wangning Cai: The University of British Columbia
Yanmei Lin: Sun Yat-sen University
Wei Ke: Sun Yat-sen University
Mustafa Hammood: The University of British Columbia
Tianye Wang: The University of British Columbia
Mengyue Xu: Sun Yat-sen University
Zibo Zheng: Université Laval
Mohammed Al-Qadasi: The University of British Columbia
Omid Esmaeeli: The University of British Columbia
Mohamed Rahim: National Research Council
Grzegorz Pakulski: National Research Council
Jens Schmid: National Research Council
Pedro Barrios: National Research Council
Weihong Jiang: National Research Council
Hugh Morison: Queen’s University
Matthew Mitchell: The University of British Columbia
Xun Guan: Tsinghua University
Nicolas A. F. Jaeger: The University of British Columbia
Leslie A. Rusch: Université Laval
Sudip Shekhar: The University of British Columbia
Wei Shi: Université Laval
Siyuan Yu: Sun Yat-sen University
Xinlun Cai: Sun Yat-sen University
Lukas Chrostowski: The University of British Columbia

Nature Communications, 2024, vol. 15, issue 1, 1-10

Abstract: Abstract Photonics offers a transformative approach to artificial intelligence (AI) and neuromorphic computing by enabling low-latency, high-speed, and energy-efficient computations. However, conventional photonic tensor cores face significant challenges in constructing large-scale photonic neuromorphic networks. Here, we propose a fully integrated photonic tensor core, consisting of only two thin-film lithium niobate (TFLN) modulators, a III-V laser, and a charge-integration photoreceiver. Despite its simple architecture, it is capable of implementing an entire layer of a neural network with a computational speed of 120 GOPS, while also allowing flexible adjustment of the number of inputs (fan-in) and outputs (fan-out). Our tensor core supports rapid in-situ training with a weight update speed of 60 GHz. Furthermore, it successfully classifies (supervised learning) and clusters (unsupervised learning) 112 × 112-pixel images through in-situ training. To enable in-situ training for clustering AI tasks, we offer a solution for performing multiplications between two negative numbers.

Date: 2024
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.nature.com/articles/s41467-024-53261-x Abstract (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:nat:natcom:v:15:y:2024:i:1:d:10.1038_s41467-024-53261-x

Ordering information: This journal article can be ordered from
https://www.nature.com/ncomms/

DOI: 10.1038/s41467-024-53261-x

Access Statistics for this article

Nature Communications is currently edited by Nathalie Le Bot, Enda Bergin and Fiona Gillespie

More articles in Nature Communications from Nature
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().