Structural plasticity-based hydrogel optical Willshaw model for one-shot on-the-fly edge learning

Wang, Dingchen1,2; Liu, Dingyao1; Lin, Yinan1; Yuan, Anran3; Zhang, Woyu4; Zhao, Yaping1,2; Wang, Shaocong1,2; Chen, Xi1,2; Chen, Hegan1,2; Zhang, Yi1,2; Jiang, Yang1,2; Shi, Shuhui1,2; Loong, Kam Chi1,2; Chen, Jia2; Wei, Songrui5; Wang, Qing6; Yu, Hongyu6; Xu, Renjing7; Shang, Dashan4; Zhang, Han5; Zhang, Shiming1; Wang, Zhongrui1,2

2023

10.1002/inf2.12399

INFOMAT   影响因子和分区

2567-3165

Autonomous one-shot on-the-fly learning copes with the high privacy, small dataset, and in-stream data at the edge. Implementing such learning on digital hardware suffers from the well-known von-Neumann and scaling bottlenecks. The optical neural networks featuring large parallelism, low latency, and high efficiency offer a promising solution. However, ex-situ training of conventional optical networks, where optical path configuration and deep learning model optimization are separated, incurs hardware, energy and time overheads, and defeats the advantages in edge learning. Here, we introduced a bio-inspired material-algorithm co-design to construct a hydrogel-based optical Willshaw model (HOWM), manifesting Hebbian-rule-based structural plasticity for simultaneous optical path configuration and deep learning model optimization thanks to the underlying opto-chemical reactions. We first employed the HOWM as an all optical in-sensor AI processor for one-shot pattern classification, association and denoising. We then leveraged HOWM to function as a ternary content addressable memory (TCAM) of an optical memory augmented neural network (MANN) for one-shot learning the Omniglot dataset. The HOWM empowered one-shot on-the-fly edge learning leads to 1000x boost of energy efficiency and 10x boost of speed, which paves the way for the next-generation autonomous, efficient, and affordable smart edge systems.

associative memory hydrogel one-shot learning optical neural network structural plasticity Willshaw model

SCI

英语

其他

National Key R&D Program of China[2018YFA0701500] ; Hong Kong Research Grant Council["27206321","17205922"] ; National Natural Science Foundation of China["62122004","61874138","61888102","61771176","62171173"] ; Chinese Academy of Sciences["XDB44000000","JCYJ20210324120409025","HZQB-KCZYZ-2021052"] ; Key Project of Department of Education of Guangdong Province[2018KCXTD026]

Materials Science

Materials Science, Multidisciplinary

WOS:000919827800001

WILEY

Web of Science

1.Univ Hong Kong, Dept Elect & Elect Engn, Hong Kong, Peoples R China
2.ACCESS AI Chip Ctr Emerging Smart Syst, InnoHK Ctr, Hong Kong Sci Pk, Hong Kong, Peoples R China
3.Macau Univ Sci & Technol, Fac Innovat Engn, Sch Comp Sci & Engn, Taipa, Macau, Peoples R China
4.Chinese Acad Sci, Inst Microelect, Key Lab Microelect Devices & Integrated Technol, Beijing, Peoples R China
5.Shenzhen Univ, Inst Microscale Optoelect, Collaborat Innovat Ctr Optoelect Sci Technol,Minis, Int Collaborat Lab Mat Optoelect 2D Mat Optoelect, Shenzhen, Peoples R China
6.Southern Univ Sci & Technol, Sch Microelect, Shenzhen, Peoples R China
7.Funct Hub Hong Kong Univ Sci & Technol Guangzhou, Microelect Thrust, Peoples Republ China, Guangzhou, Guagndong, Peoples R China

Wang, Dingchen,Liu, Dingyao,Lin, Yinan,et al. Structural plasticity-based hydrogel optical Willshaw model for one-shot on-the-fly edge learning[J]. INFOMAT,2023,5(4).

Wang, Dingchen.,Liu, Dingyao.,Lin, Yinan.,Yuan, Anran.,Zhang, Woyu.,...&Wang, Zhongrui.(2023).Structural plasticity-based hydrogel optical Willshaw model for one-shot on-the-fly edge learning.INFOMAT,5(4).

Wang, Dingchen,et al."Structural plasticity-based hydrogel optical Willshaw model for one-shot on-the-fly edge learning".INFOMAT 5.4(2023).