Cargando…

Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy

Neural networks based on memristive devices have achieved great progress recently. However, memristive synapses with nonlinearity and asymmetry seriously limit the classification accuracy. Moreover, insufficient number of training samples in many cases also have negative effect on the classification...

Descripción completa

Detalles Bibliográficos
Autores principales: Huang, He‐Ming, Xiao, Yu, Yang, Rui, Yu, Ye‐Tian, He, Hui‐Kai, Wang, Zhe, Guo, Xin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: John Wiley and Sons Inc. 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7509653/
https://www.ncbi.nlm.nih.gov/pubmed/32999852
http://dx.doi.org/10.1002/advs.202001842
_version_ 1783585642202005504
author Huang, He‐Ming
Xiao, Yu
Yang, Rui
Yu, Ye‐Tian
He, Hui‐Kai
Wang, Zhe
Guo, Xin
author_facet Huang, He‐Ming
Xiao, Yu
Yang, Rui
Yu, Ye‐Tian
He, Hui‐Kai
Wang, Zhe
Guo, Xin
author_sort Huang, He‐Ming
collection PubMed
description Neural networks based on memristive devices have achieved great progress recently. However, memristive synapses with nonlinearity and asymmetry seriously limit the classification accuracy. Moreover, insufficient number of training samples in many cases also have negative effect on the classification accuracy of neural networks due to overfitting. In this work, dropout neuronal units are developed based on stochastic volatile memristive devices of Ag/Ta(2)O(5):Ag/Pt. The memristive neural network using the dropout neuronal units effectively solves the problem of overfitting and mitigates the negative effects of the nonideality of memristive synapses, eventually achieves a classification accuracy comparable to the theoretical limit. The stochastic and volatile switching performances of the Ag/Ta(2)O(5):Ag/Pt device are attributed to the stochastical rupture of the Ag filament under high electrical stress in the Ta(2)O(5) layer, according to the TEM observation and the kinetic Monte Carlo simulation.
format Online
Article
Text
id pubmed-7509653
institution National Center for Biotechnology Information
language English
publishDate 2020
publisher John Wiley and Sons Inc.
record_format MEDLINE/PubMed
spelling pubmed-75096532020-09-29 Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy Huang, He‐Ming Xiao, Yu Yang, Rui Yu, Ye‐Tian He, Hui‐Kai Wang, Zhe Guo, Xin Adv Sci (Weinh) Communications Neural networks based on memristive devices have achieved great progress recently. However, memristive synapses with nonlinearity and asymmetry seriously limit the classification accuracy. Moreover, insufficient number of training samples in many cases also have negative effect on the classification accuracy of neural networks due to overfitting. In this work, dropout neuronal units are developed based on stochastic volatile memristive devices of Ag/Ta(2)O(5):Ag/Pt. The memristive neural network using the dropout neuronal units effectively solves the problem of overfitting and mitigates the negative effects of the nonideality of memristive synapses, eventually achieves a classification accuracy comparable to the theoretical limit. The stochastic and volatile switching performances of the Ag/Ta(2)O(5):Ag/Pt device are attributed to the stochastical rupture of the Ag filament under high electrical stress in the Ta(2)O(5) layer, according to the TEM observation and the kinetic Monte Carlo simulation. John Wiley and Sons Inc. 2020-07-26 /pmc/articles/PMC7509653/ /pubmed/32999852 http://dx.doi.org/10.1002/advs.202001842 Text en © 2020 The Authors. Published by WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim This is an open access article under the terms of the http://creativecommons.org/licenses/by/4.0/ License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
spellingShingle Communications
Huang, He‐Ming
Xiao, Yu
Yang, Rui
Yu, Ye‐Tian
He, Hui‐Kai
Wang, Zhe
Guo, Xin
Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title_full Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title_fullStr Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title_full_unstemmed Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title_short Implementation of Dropout Neuronal Units Based on Stochastic Memristive Devices in Neural Networks with High Classification Accuracy
title_sort implementation of dropout neuronal units based on stochastic memristive devices in neural networks with high classification accuracy
topic Communications
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7509653/
https://www.ncbi.nlm.nih.gov/pubmed/32999852
http://dx.doi.org/10.1002/advs.202001842
work_keys_str_mv AT huangheming implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT xiaoyu implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT yangrui implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT yuyetian implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT hehuikai implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT wangzhe implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy
AT guoxin implementationofdropoutneuronalunitsbasedonstochasticmemristivedevicesinneuralnetworkswithhighclassificationaccuracy