RRAM based convolutional neural networks for high accuracy pattern recognition and online learning tasks

Zhen Dong, Z. Zhou, Z.F. Li, C. Liu, Y. Jiang, P. Huang, L.F. Liu, X.Y. Liu, J. Kang
{"title":"RRAM based convolutional neural networks for high accuracy pattern recognition and online learning tasks","authors":"Zhen Dong, Z. Zhou, Z.F. Li, C. Liu, Y. Jiang, P. Huang, L.F. Liu, X.Y. Liu, J. Kang","doi":"10.23919/SNW.2017.8242339","DOIUrl":null,"url":null,"abstract":"In this work, we conduct research on optimizing schemes for the RRAM-based implementation of CNN. Our main achievements contain: 1) A concrete CNN circuit and corresponding operation methods are developed. 2) Quantification methods for utilizing binary or multilevel RRAM as synapses are proposed, and our CNN performs with 98% accuracy on the MNIST dataset using multilevel RRAM and 97% accuracy using binary RRAM. 3) Influence of the number and size of kernels, as well as the device conductance variation on final recognition accuracy is studied in detail. 4) Online learning tasks are performed using the developed CNN system with binary STDP protocol, and 94.6% accuracy on average is achieved.","PeriodicalId":424135,"journal":{"name":"2017 Silicon Nanoelectronics Workshop (SNW)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 Silicon Nanoelectronics Workshop (SNW)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/SNW.2017.8242339","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 11

Abstract

In this work, we conduct research on optimizing schemes for the RRAM-based implementation of CNN. Our main achievements contain: 1) A concrete CNN circuit and corresponding operation methods are developed. 2) Quantification methods for utilizing binary or multilevel RRAM as synapses are proposed, and our CNN performs with 98% accuracy on the MNIST dataset using multilevel RRAM and 97% accuracy using binary RRAM. 3) Influence of the number and size of kernels, as well as the device conductance variation on final recognition accuracy is studied in detail. 4) Online learning tasks are performed using the developed CNN system with binary STDP protocol, and 94.6% accuracy on average is achieved.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
基于RRAM的卷积神经网络用于高精度模式识别和在线学习任务
在这项工作中,我们对基于ram的CNN实现的优化方案进行了研究。我们的主要成果包括:1)开发了一个具体的CNN电路和相应的操作方法。2)提出了使用二进制或多层RRAM作为突触的量化方法,我们的CNN在使用多层RRAM的MNIST数据集上的准确率为98%,使用二进制RRAM的准确率为97%。3)详细研究了核数、核尺寸以及器件电导变化对最终识别精度的影响。4)使用开发的CNN系统,采用二进制STDP协议进行在线学习任务,平均准确率达到94.6%。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Revisiting room-temperature 1.54 μιη photoluminescence of ErOx centers in silicon at extremely low concentration Undoped SiGe FETs with metal-insulator-semiconductor contacts Improved electrical characteristics and reliability of multi-stacking PNPN junctionless transistors using channel depletion effect Program/erase speed and data retention trade-off in negative capacitance versatile memory Investigations on dynamic characteristics of ferroelectric Hf02 based on multi-domain interaction model
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1