INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY |
Prev
Next
|
|
|
One memristor-one electrolyte-gated transistor-based high energy-efficient dropout neuronal units |
Yalin Li(李亚霖)1,†, Kailu Shi(时凯璐)1,†, Yixin Zhu(朱一新)1, Xiao Fang(方晓)1, Hangyuan Cui(崔航源)1, Qing Wan(万青)2,‡, and Changjin Wan(万昌锦)1,§ |
1 School of Electronic Science & Engineering, and Collaborative Innovation Center of Advanced Microstructures, Nanjing University, Nanjing 210023, China; 2 Yongjiang Laboratory (Y-LAB), Ningbo 315202, China |
|
|
Abstract Artificial neural networks (ANN) have been extensively researched due to their significant energy-saving benefits. Hardware implementations of ANN with dropout function would be able to avoid the overfitting problem. This letter reports a dropout neuronal unit (1R1T-DNU) based on one memristor-one electrolyte-gated transistor with an ultralow energy consumption of 25pJ/spike. A dropout neural network is constructed based on such a device and has been verified by MNIST dataset, demonstrating high recognition accuracies ($> 90$%) within a large range of dropout probabilities up to 40%. The running time can be reduced by increasing dropout probability without a significant loss in accuracy. Our results indicate the great potential of introducing such 1R1T-DNUs in full-hardware neural networks to enhance energy efficiency and to solve the overfitting problem.
|
Received: 08 March 2024
Revised: 19 March 2024
Accepted manuscript online: 03 April 2024
|
PACS:
|
84.35.+i
|
(Neural networks)
|
|
73.40.Mr
|
(Semiconductor-electrolyte contacts)
|
|
77.55.-g
|
(Dielectric thin films)
|
|
81.05.Gc
|
(Amorphous semiconductors)
|
|
Fund: Project supported by the National Key Research and Development Program of China (Grant Nos. 2021YFA1202600 and 2023YFE0208600) and in part by the National Natural Science Foundation of China (Grant Nos. 62174082, 92364106, 61921005, 92364204, and 62074075). |
Corresponding Authors:
Qing Wan, Changjin Wan
E-mail: wanqing@nju.edu.cn;cjwan@nju.edu.cn
|
Cite this article:
Yalin Li(李亚霖), Kailu Shi(时凯璐), Yixin Zhu(朱一新), Xiao Fang(方晓), Hangyuan Cui(崔航源), Qing Wan(万青), and Changjin Wan(万昌锦) One memristor-one electrolyte-gated transistor-based high energy-efficient dropout neuronal units 2024 Chin. Phys. B 33 068401
|
[1] López C 2023 Advanced Materials 35 2208683 [2] Prezioso M, Merrikh-Bayat F, Hoskins B D, Adam G C, Likharev K K and Strukov D B 2015 Nature 521 61 [3] Ahn E C 2020 npj 2D Materials and Applications 4 17 [4] Sun K, Chen J and Yan X 2021 Adv. Funct. Mater. 31 2006773 [5] Zhu L Q, Wan C J, Guo L Q, Shi Y and Wan Q 2014 Nat. Commun. 5 3158 [6] Zhu Y, He Y, Chen C, Zhu L, Mao H, Zhu Y, Wang X, Yang Y, Wan C and Wan Q 2022 Appl. Phys. Lett. 120 113504 [7] Zhu Y, Peng B, Zhu L, Chen C, Wang X, Mao H, Zhu Y, Fu C, Ke S, Wan C and Wan Q 2022 Appl. Phys. Lett. 121 133502 [8] Lin X, Long H, Ke S, Wang Y, Zhu Y, Chen C, Wan C and Wan Q 2022 Chin. Phys. Lett. 39 068501 [9] Wu X, Zhang Y, Zhang T and Zhang B 2023 Chin. Phys. B 32 100506 [10] Xiang T, Chen F, Li X, Wang X, Yan Y and Wang L 2023 Chin. Phys. B 32 117301 [11] Yang Y, Fu C, Ke S, Cui H, Fang X, Wan C and Wan Q 2023 Chin. Phys. B 32 118101 [12] Amirsoleimani A, Alibart F, Yon V, Xu J, Pazhouhandeh M R, Ecoffey S, Beilliard Y, Genov R and Drouin D 2020 Advanced Intelligent Systems 2 2000115 [13] Demasius K U, Kirschen A and Parkin S 2021 Nat. Electron. 4 748 [14] Srivastava N, Hinton G, Krizhevsky A, Sutskever I and Salakhutdinov R 2014 Journal of Machine Learning Research 15 1929 [15] Huang H M, Xiao Y, Yang R, Yu Y T, He H K, Wang Z and Guo X 2020 Adv. Sci. 7 2001842 [16] Wang Z, Joshi S, Savel’ev S E, Jiang H, Midya R, Lin P, Hu M, Ge N, Strachan J P, Li Z, Wu Q, Barnell M, Li G L, Xin H L, Williams R S, Xia Q and Yang J J 2017 Nat. Mater. 16 101 [17] Wang Z, Wu H, Burr G W, Hwang C S, Wang K L, Xia Q and Yang J J 2020 Nat. Rev. Mater. 5 173 [18] Wan C, Zhou J, Shi Y and Wan Q 2014 IEEE Electron Dev. Lett. 35 414 [19] Wan X, Feng P, Wu G D, Shi Y and Wan Q 2015 IEEE Electron Dev. Lett. 36 204 [20] Shao F, Feng P, Wan C, Wan X, Yang Y, Shi Y and Wan Q 2017 Adv. Electron. Mater. 3 1600509 [21] Huang Q, Huang R, Pan Y, Tan S and Wang Y 2014 IEEE Electron Dev. Lett. 35 877 [22] Mazady A and Anwar M 2014 IEEE Trans. Electron Dev. 61 1054 [23] Sanjar K, Rehman A, Paul A and JeongHong K 2020 8th International Conference on Orange Technology (ICOT), 18-21 December, 2020, pp. 1-4 [24] Zhu Y, Peng B, Zhu L, Chen C, Wang X, Mao H, Zhu Y, Fu C, Ke S, Wan C and Wan Q 2022 Appl. Phys. Lett. 121 133502 [25] Mendenhall J and Meiler J 2016 Journal of Computer-Aided Molecular Design 30 177 |
No Suggested Reading articles found! |
|
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
Altmetric
|
blogs
Facebook pages
Wikipedia page
Google+ users
|
Online attention
Altmetric calculates a score based on the online attention an article receives. Each coloured thread in the circle represents a different type of online attention. The number in the centre is the Altmetric score. Social media and mainstream news media are the main sources that calculate the score. Reference managers such as Mendeley are also tracked but do not contribute to the score. Older articles often score higher because they have had more time to get noticed. To account for this, Altmetric has included the context data for other articles of a similar age.
View more on Altmetrics
|
|
|