Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning

Christopher H. Bennett; T. Patrick Xiao; Can Cui; Naimul Hassan; Otitoaleke G. Akinola; Jean Anne C. Incorvia; Alvaro Velasquez; Joseph S. Friedman; Matthew J. Marinella

Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning

Christopher H. Bennett
, T. Patrick Xiao
, Can Cui
, Naimul Hassan
, Otitoaleke G. Akinola
, Jean Anne C. Incorvia
, Alvaro Velasquez
, Joseph S. Friedman
, Matthew J. Marinella

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

Abstract

Machine learning implements backpropagation via abundant training samples. We demonstrate a multi-stage learning system realized by a promising non-volatile memory device, the domain-wall magnetic tunnel junction (DW-MTJ). The system consists of unsupervised (clustering) as well as supervised sub-systems, and generalizes quickly (with few samples). We demonstrate interactions between physical properties of this device and optimal implementation of neuroscience-inspired plasticity learning rules, and highlight performance on a suite of tasks. Our energy analysis confirms the value of the approach, as the learning budget stays below 20µJ even for large tasks used typically in machine learning.

Original language	English (US)
Title of host publication	2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728133201
State	Published - 2020
Externally published	Yes
Event	52nd IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Virtual, Online Duration: Oct 10 2020 → Oct 21 2020

Publication series

Name	Proceedings - IEEE International Symposium on Circuits and Systems
Volume	2020-October

Conference

Conference	52nd IEEE International Symposium on Circuits and Systems, ISCAS 2020
City	Virtual, Online
Period	10/10/20 → 10/21/20

ASJC Scopus subject areas

Electrical and Electronic Engineering

Cite this

Bennett, C. H., Xiao, T. P., Cui, C., Hassan, N., Akinola, O. G., Incorvia, J. A. C., Velasquez, A., Friedman, J. S., & Marinella, M. J. (2020). Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning. In 2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings Article 9180511 (Proceedings - IEEE International Symposium on Circuits and Systems; Vol. 2020-October). Institute of Electrical and Electronics Engineers Inc..

Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning. / Bennett, Christopher H.; Xiao, T. Patrick; Cui, Can et al.
2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2020. 9180511 (Proceedings - IEEE International Symposium on Circuits and Systems; Vol. 2020-October).

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

Bennett, CH, Xiao, TP, Cui, C, Hassan, N, Akinola, OG, Incorvia, JAC, Velasquez, A, Friedman, JS & Marinella, MJ 2020, Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning. in 2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings., 9180511, Proceedings - IEEE International Symposium on Circuits and Systems, vol. 2020-October, Institute of Electrical and Electronics Engineers Inc., 52nd IEEE International Symposium on Circuits and Systems, ISCAS 2020, Virtual, Online, 10/10/20.

Bennett CH, Xiao TP, Cui C, Hassan N, Akinola OG, Incorvia JAC et al. Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning. In 2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2020. 9180511. (Proceedings - IEEE International Symposium on Circuits and Systems).

@inproceedings{c040814acba04de6b21d241401e1af74,

title = "Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning",

abstract = "Machine learning implements backpropagation via abundant training samples. We demonstrate a multi-stage learning system realized by a promising non-volatile memory device, the domain-wall magnetic tunnel junction (DW-MTJ). The system consists of unsupervised (clustering) as well as supervised sub-systems, and generalizes quickly (with few samples). We demonstrate interactions between physical properties of this device and optimal implementation of neuroscience-inspired plasticity learning rules, and highlight performance on a suite of tasks. Our energy analysis confirms the value of the approach, as the learning budget stays below 20µJ even for large tasks used typically in machine learning.",

author = "Bennett, \{Christopher H.\} and Xiao, \{T. Patrick\} and Can Cui and Naimul Hassan and Akinola, \{Otitoaleke G.\} and Incorvia, \{Jean Anne C.\} and Alvaro Velasquez and Friedman, \{Joseph S.\} and Marinella, \{Matthew J.\}",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE; 52nd IEEE International Symposium on Circuits and Systems, ISCAS 2020 ; Conference date: 10-10-2020 Through 21-10-2020",

year = "2020",

language = "English (US)",

series = "Proceedings - IEEE International Symposium on Circuits and Systems",

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

booktitle = "2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings",

}

TY - GEN

T1 - Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning

AU - Bennett, Christopher H.

AU - Xiao, T. Patrick

AU - Cui, Can

AU - Hassan, Naimul

AU - Akinola, Otitoaleke G.

AU - Incorvia, Jean Anne C.

AU - Velasquez, Alvaro

AU - Friedman, Joseph S.

AU - Marinella, Matthew J.

PY - 2020

Y1 - 2020

N2 - Machine learning implements backpropagation via abundant training samples. We demonstrate a multi-stage learning system realized by a promising non-volatile memory device, the domain-wall magnetic tunnel junction (DW-MTJ). The system consists of unsupervised (clustering) as well as supervised sub-systems, and generalizes quickly (with few samples). We demonstrate interactions between physical properties of this device and optimal implementation of neuroscience-inspired plasticity learning rules, and highlight performance on a suite of tasks. Our energy analysis confirms the value of the approach, as the learning budget stays below 20µJ even for large tasks used typically in machine learning.

AB - Machine learning implements backpropagation via abundant training samples. We demonstrate a multi-stage learning system realized by a promising non-volatile memory device, the domain-wall magnetic tunnel junction (DW-MTJ). The system consists of unsupervised (clustering) as well as supervised sub-systems, and generalizes quickly (with few samples). We demonstrate interactions between physical properties of this device and optimal implementation of neuroscience-inspired plasticity learning rules, and highlight performance on a suite of tasks. Our energy analysis confirms the value of the approach, as the learning budget stays below 20µJ even for large tasks used typically in machine learning.

UR - https://www.scopus.com/pages/publications/85109350254

UR - https://www.scopus.com/pages/publications/85109350254#tab=citedBy

M3 - Conference contribution

T3 - Proceedings - IEEE International Symposium on Circuits and Systems

BT - 2020 IEEE International Symposium on Circuits and Systems, ISCAS 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 52nd IEEE International Symposium on Circuits and Systems, ISCAS 2020

Y2 - 10 October 2020 through 21 October 2020

ER -

Plasticity-enhanced domain-wall MTJ neural networks for energy-efficient online learning

Abstract

Publication series

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this