首页|One memristor-one electrolyte-gated transistor-based high energy-efficient dropout neuronal units

One memristor-one electrolyte-gated transistor-based high energy-efficient dropout neuronal units

扫码查看
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 25 pJ/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 IRIT-DNUs in full-hardware neural networks to enhance energy efficiency and to solve the overfitting problem.

dropout neuronal unitsynaptic transistorsmemristorartificial neural network

李亚霖、时凯璐、朱一新、方晓、崔航源、万青、万昌锦

展开 >

School of Electronic Science & Engineering,and Collaborative Innovation Center of Advanced Microstructures,Nanjing University,Nanjing 210023,China

Yongjiang Laboratory(Y-LAB),Ningbo 315202,China

国家重点研发计划国家重点研发计划国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金国家自然科学基金

2021YFA12026002023YFE02086006217408292364106619210059236420462074075

2024

中国物理B(英文版)
中国物理学会和中国科学院物理研究所

中国物理B(英文版)

CSTPCDEI
影响因子:0.995
ISSN:1674-1056
年,卷(期):2024.33(6)
  • 25