Nature Communications (Oct 2024)
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,
- Lukas Chrostowski
Affiliations
- Zhongjin Lin
- Department of Electrical and Computer Engineering, The University of British Columbia
- Bhavin J. Shastri
- Department of Physics, Engineering Physics and Astronomy, Queen’s University
- Shangxuan Yu
- Department of Electrical and Computer Engineering, The University of British Columbia
- Jingxiang Song
- Department of Electrical and Computer Engineering, The University of British Columbia
- Yuntao Zhu
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Arman Safarnejadian
- Department of Electrical and Computer Engineering, Université Laval
- Wangning Cai
- Department of Electrical and Computer Engineering, The University of British Columbia
- Yanmei Lin
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Wei Ke
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Mustafa Hammood
- Department of Electrical and Computer Engineering, The University of British Columbia
- Tianye Wang
- Department of Electrical and Computer Engineering, The University of British Columbia
- Mengyue Xu
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Zibo Zheng
- Department of Electrical and Computer Engineering, Université Laval
- Mohammed Al-Qadasi
- Department of Electrical and Computer Engineering, The University of British Columbia
- Omid Esmaeeli
- Department of Electrical and Computer Engineering, The University of British Columbia
- Mohamed Rahim
- Advanced Electronics and Photonics Research Centre, National Research Council
- Grzegorz Pakulski
- Advanced Electronics and Photonics Research Centre, National Research Council
- Jens Schmid
- Advanced Electronics and Photonics Research Centre, National Research Council
- Pedro Barrios
- Advanced Electronics and Photonics Research Centre, National Research Council
- Weihong Jiang
- Advanced Electronics and Photonics Research Centre, National Research Council
- Hugh Morison
- Department of Physics, Engineering Physics and Astronomy, Queen’s University
- Matthew Mitchell
- Department of Electrical and Computer Engineering, The University of British Columbia
- Xun Guan
- Tsinghua Shenzhen International Graduate School, Tsinghua University
- Nicolas A. F. Jaeger
- Department of Electrical and Computer Engineering, The University of British Columbia
- Leslie A. Rusch
- Department of Electrical and Computer Engineering, Université Laval
- Sudip Shekhar
- Department of Electrical and Computer Engineering, The University of British Columbia
- Wei Shi
- Department of Electrical and Computer Engineering, Université Laval
- Siyuan Yu
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Xinlun Cai
- State Key Laboratory of Optoelectronic Materials and Technologies, School of Electronics and Information Technology, Sun Yat-sen University
- Lukas Chrostowski
- Department of Electrical and Computer Engineering, The University of British Columbia
- DOI
- https://doi.org/10.1038/s41467-024-53261-x
- Journal volume & issue
-
Vol. 15,
no. 1
pp. 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.
