Majority neuron circuit having large fan-in with non-volatile synaptic weight

Hisanao Akima; Yasuhiro Katayama; Koji Nakajima; Masao Sakuraba; Shigeo Sato

doi:10.1109/IJCNN.2014.6889766

Majority neuron circuit having large fan-in with non-volatile synaptic weight

Hisanao Akima, Yasuhiro Katayama, Koji Nakajima, Masao Sakuraba, Shigeo Sato

RIEC - Computing System Platforms Division

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

1 Citation (Scopus)

Abstract

We present a design of a majority neuron circuit with non-volatile synaptic weights. It is based on an analog majority circuit composed of controlled current inverters (CCIs). The proposed circuit is immune to device parameter fluctuations, and its fan-in is estimated about 1000. Synaptic weights are realized on the neuron circuit by adding variable resistors. We consider a design of a non-volatile synaptic weight by using a three-terminal magnetic domain-wall motion (DWM) device. The operation of a fully connected recurrent neural network composed of the proposed circuits has been confirmed by SPICE simulation.

Original language	English
Title of host publication	Proceedings of the International Joint Conference on Neural Networks
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	4266-4271
Number of pages	6
ISBN (Electronic)	9781479914845
DOIs	https://doi.org/10.1109/IJCNN.2014.6889766
Publication status	Published - 2014 Sept 3
Event	2014 International Joint Conference on Neural Networks, IJCNN 2014 - Beijing, China Duration: 2014 Jul 6 → 2014 Jul 11

Publication series

Name	Proceedings of the International Joint Conference on Neural Networks

Conference

Conference	2014 International Joint Conference on Neural Networks, IJCNN 2014
Country/Territory	China
City	Beijing
Period	14/7/6 → 14/7/11

Access to Document

10.1109/IJCNN.2014.6889766

Cite this

Akima, H., Katayama, Y., Nakajima, K., Sakuraba, M., & Sato, S. (2014). Majority neuron circuit having large fan-in with non-volatile synaptic weight. In Proceedings of the International Joint Conference on Neural Networks (pp. 4266-4271). Article 6889766 (Proceedings of the International Joint Conference on Neural Networks). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.2014.6889766

@inproceedings{b9fa5841e1a344798027d221760cac7d,

title = "Majority neuron circuit having large fan-in with non-volatile synaptic weight",

abstract = "We present a design of a majority neuron circuit with non-volatile synaptic weights. It is based on an analog majority circuit composed of controlled current inverters (CCIs). The proposed circuit is immune to device parameter fluctuations, and its fan-in is estimated about 1000. Synaptic weights are realized on the neuron circuit by adding variable resistors. We consider a design of a non-volatile synaptic weight by using a three-terminal magnetic domain-wall motion (DWM) device. The operation of a fully connected recurrent neural network composed of the proposed circuits has been confirmed by SPICE simulation.",

author = "Hisanao Akima and Yasuhiro Katayama and Koji Nakajima and Masao Sakuraba and Shigeo Sato",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 International Joint Conference on Neural Networks, IJCNN 2014 ; Conference date: 06-07-2014 Through 11-07-2014",

year = "2014",

month = sep,

day = "3",

doi = "10.1109/IJCNN.2014.6889766",

language = "English",

series = "Proceedings of the International Joint Conference on Neural Networks",

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

pages = "4266--4271",

booktitle = "Proceedings of the International Joint Conference on Neural Networks",

address = "United States",

}

Akima, H, Katayama, Y, Nakajima, K, Sakuraba, M & Sato, S 2014, Majority neuron circuit having large fan-in with non-volatile synaptic weight. in Proceedings of the International Joint Conference on Neural Networks., 6889766, Proceedings of the International Joint Conference on Neural Networks, Institute of Electrical and Electronics Engineers Inc., pp. 4266-4271, 2014 International Joint Conference on Neural Networks, IJCNN 2014, Beijing, China, 14/7/6. https://doi.org/10.1109/IJCNN.2014.6889766

Majority neuron circuit having large fan-in with non-volatile synaptic weight. / Akima, Hisanao; Katayama, Yasuhiro; Nakajima, Koji et al.
Proceedings of the International Joint Conference on Neural Networks. Institute of Electrical and Electronics Engineers Inc., 2014. p. 4266-4271 6889766 (Proceedings of the International Joint Conference on Neural Networks).

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

TY - GEN

T1 - Majority neuron circuit having large fan-in with non-volatile synaptic weight

AU - Akima, Hisanao

AU - Katayama, Yasuhiro

AU - Nakajima, Koji

AU - Sakuraba, Masao

AU - Sato, Shigeo

PY - 2014/9/3

Y1 - 2014/9/3

N2 - We present a design of a majority neuron circuit with non-volatile synaptic weights. It is based on an analog majority circuit composed of controlled current inverters (CCIs). The proposed circuit is immune to device parameter fluctuations, and its fan-in is estimated about 1000. Synaptic weights are realized on the neuron circuit by adding variable resistors. We consider a design of a non-volatile synaptic weight by using a three-terminal magnetic domain-wall motion (DWM) device. The operation of a fully connected recurrent neural network composed of the proposed circuits has been confirmed by SPICE simulation.

AB - We present a design of a majority neuron circuit with non-volatile synaptic weights. It is based on an analog majority circuit composed of controlled current inverters (CCIs). The proposed circuit is immune to device parameter fluctuations, and its fan-in is estimated about 1000. Synaptic weights are realized on the neuron circuit by adding variable resistors. We consider a design of a non-volatile synaptic weight by using a three-terminal magnetic domain-wall motion (DWM) device. The operation of a fully connected recurrent neural network composed of the proposed circuits has been confirmed by SPICE simulation.

UR - http://www.scopus.com/inward/record.url?scp=84908474227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84908474227&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2014.6889766

DO - 10.1109/IJCNN.2014.6889766

M3 - Conference contribution

AN - SCOPUS:84908474227

T3 - Proceedings of the International Joint Conference on Neural Networks

SP - 4266

EP - 4271

BT - Proceedings of the International Joint Conference on Neural Networks

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 International Joint Conference on Neural Networks, IJCNN 2014

Y2 - 6 July 2014 through 11 July 2014

ER -

Akima H, Katayama Y, Nakajima K, Sakuraba M , Sato S. Majority neuron circuit having large fan-in with non-volatile synaptic weight. In Proceedings of the International Joint Conference on Neural Networks. Institute of Electrical and Electronics Engineers Inc. 2014. p. 4266-4271. 6889766. (Proceedings of the International Joint Conference on Neural Networks). doi: 10.1109/IJCNN.2014.6889766

Majority neuron circuit having large fan-in with non-volatile synaptic weight

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this