Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization

Peng Yao; Huaqiang Wu; Bin Gao; Ning Deng; Shimeng Yu; He Qian

doi:10.1109/EDTM.2017.7947592

Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization

Peng Yao, Huaqiang Wu, Bin Gao, Ning Deng, Shimeng Yu, He Qian

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

4 Scopus citations

Abstract

In this work, online training is experimentally demonstrated on a neuromorphic network built with 1k-bit 1T1R RRAM array. The 1T1R RRAM cells in the array exhibit excellent synaptic behavior. Patterns with up to 11.83% noises can be correctly classified by the network after training. Based on the analysis of experimental results, we find the device characteristics and operation schemes significantly affect the system performance. The impact of voltage dependence and asymmetry effects are evaluated, and optimization guidelines are provided.

Original language	English (US)
Title of host publication	2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	182-183
Number of pages	2
ISBN (Electronic)	9781509046591
DOIs	https://doi.org/10.1109/EDTM.2017.7947592
State	Published - Jun 13 2017
Event	2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Toyama, Japan Duration: Feb 28 2017 → Mar 2 2017

Other

Other	2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017
Country	Japan
City	Toyama
Period	2/28/17 → 3/2/17

Keywords

network
RRAM
synaptic behavior

ASJC Scopus subject areas

Hardware and Architecture
Electrical and Electronic Engineering

Access to Document

10.1109/EDTM.2017.7947592

Fingerprint Dive into the research topics of 'Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization'. Together they form a unique fingerprint.

Cite this

Yao, P., Wu, H., Gao, B., Deng, N., Yu, S., & Qian, H. (2017). Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization. In 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings (pp. 182-183). [7947592] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/EDTM.2017.7947592

Online training on RRAM based neuromorphic network : Experimental demonstration and operation scheme optimization. / Yao, Peng; Wu, Huaqiang; Gao, Bin; Deng, Ning; Yu, Shimeng; Qian, He.

2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 182-183 7947592.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yao, P, Wu, H, Gao, B, Deng, N, Yu, S & Qian, H 2017, Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization. in 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings., 7947592, Institute of Electrical and Electronics Engineers Inc., pp. 182-183, 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017, Toyama, Japan, 2/28/17. https://doi.org/10.1109/EDTM.2017.7947592

Yao P, Wu H, Gao B, Deng N, Yu S, Qian H. Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization. In 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 182-183. 7947592 https://doi.org/10.1109/EDTM.2017.7947592

@inproceedings{70b63788312d45bbadff0882fc93fd01,

title = "Online training on RRAM based neuromorphic network: Experimental demonstration and operation scheme optimization",

abstract = "In this work, online training is experimentally demonstrated on a neuromorphic network built with 1k-bit 1T1R RRAM array. The 1T1R RRAM cells in the array exhibit excellent synaptic behavior. Patterns with up to 11.83% noises can be correctly classified by the network after training. Based on the analysis of experimental results, we find the device characteristics and operation schemes significantly affect the system performance. The impact of voltage dependence and asymmetry effects are evaluated, and optimization guidelines are provided.",

keywords = "network, RRAM, synaptic behavior",

author = "Peng Yao and Huaqiang Wu and Bin Gao and Ning Deng and Shimeng Yu and He Qian",

year = "2017",

month = jun,

day = "13",

doi = "10.1109/EDTM.2017.7947592",

language = "English (US)",

pages = "182--183",

booktitle = "2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 ; Conference date: 28-02-2017 Through 02-03-2017",

}

TY - GEN

T1 - Online training on RRAM based neuromorphic network

T2 - 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017

AU - Yao, Peng

AU - Wu, Huaqiang

AU - Gao, Bin

AU - Deng, Ning

AU - Yu, Shimeng

AU - Qian, He

PY - 2017/6/13

Y1 - 2017/6/13

N2 - In this work, online training is experimentally demonstrated on a neuromorphic network built with 1k-bit 1T1R RRAM array. The 1T1R RRAM cells in the array exhibit excellent synaptic behavior. Patterns with up to 11.83% noises can be correctly classified by the network after training. Based on the analysis of experimental results, we find the device characteristics and operation schemes significantly affect the system performance. The impact of voltage dependence and asymmetry effects are evaluated, and optimization guidelines are provided.

AB - In this work, online training is experimentally demonstrated on a neuromorphic network built with 1k-bit 1T1R RRAM array. The 1T1R RRAM cells in the array exhibit excellent synaptic behavior. Patterns with up to 11.83% noises can be correctly classified by the network after training. Based on the analysis of experimental results, we find the device characteristics and operation schemes significantly affect the system performance. The impact of voltage dependence and asymmetry effects are evaluated, and optimization guidelines are provided.

KW - network

KW - RRAM

KW - synaptic behavior

UR - http://www.scopus.com/inward/record.url?scp=85021881261&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85021881261&partnerID=8YFLogxK

U2 - 10.1109/EDTM.2017.7947592

DO - 10.1109/EDTM.2017.7947592

M3 - Conference contribution

AN - SCOPUS:85021881261

SP - 182

EP - 183

BT - 2017 IEEE Electron Devices Technology and Manufacturing Conference, EDTM 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 28 February 2017 through 2 March 2017

ER -