Power scalable implementation of artificial neural networks

S. S. Modi, P. R. Wilson, A. D. Brown

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

As the use of Artificial Neural Network(ANN) in mobile embedded devices gets more pervasive, power consumption of ANN hardware is becoming a major limiting factor. Although considerable research efforts are now directed towards low-power implementations of ANN, the issue of dynamic power scalability of the implemented design has been largely overlooked. In this paper, we discuss the motivation and basic principles for implementing power scaling in ANN Hardware. With the help of a simple example, we demonstrate how power scaling can be achieved with dynamic pruning techniques.
Original languageEnglish
Title of host publication12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005
PublisherIEEE
Pages1-4
ISBN (Print)9789972611001
DOIs
Publication statusPublished - 2005
Event12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005 - Gammarth, Tunisia
Duration: 11 Dec 200514 Dec 2005

Conference

Conference12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005
CountryTunisia
CityGammarth
Period11/12/0514/12/05

Fingerprint

Neural networks
Hardware
Scalability
Electric power utilization

Cite this

Modi, S. S., Wilson, P. R., & Brown, A. D. (2005). Power scalable implementation of artificial neural networks. In 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005 (pp. 1-4). IEEE. https://doi.org/10.1109/ICECS.2005.4633538

Power scalable implementation of artificial neural networks. / Modi, S. S.; Wilson, P. R.; Brown, A. D.

12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005. IEEE, 2005. p. 1-4.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Modi, SS, Wilson, PR & Brown, AD 2005, Power scalable implementation of artificial neural networks. in 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005. IEEE, pp. 1-4, 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005, Gammarth, Tunisia, 11/12/05. https://doi.org/10.1109/ICECS.2005.4633538
Modi SS, Wilson PR, Brown AD. Power scalable implementation of artificial neural networks. In 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005. IEEE. 2005. p. 1-4 https://doi.org/10.1109/ICECS.2005.4633538
Modi, S. S. ; Wilson, P. R. ; Brown, A. D. / Power scalable implementation of artificial neural networks. 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005. IEEE, 2005. pp. 1-4
@inproceedings{27ab852753204670ae0cb8b1d64e5341,
title = "Power scalable implementation of artificial neural networks",
abstract = "As the use of Artificial Neural Network(ANN) in mobile embedded devices gets more pervasive, power consumption of ANN hardware is becoming a major limiting factor. Although considerable research efforts are now directed towards low-power implementations of ANN, the issue of dynamic power scalability of the implemented design has been largely overlooked. In this paper, we discuss the motivation and basic principles for implementing power scaling in ANN Hardware. With the help of a simple example, we demonstrate how power scaling can be achieved with dynamic pruning techniques.",
author = "Modi, {S. S.} and Wilson, {P. R.} and Brown, {A. D.}",
year = "2005",
doi = "10.1109/ICECS.2005.4633538",
language = "English",
isbn = "9789972611001",
pages = "1--4",
booktitle = "12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005",
publisher = "IEEE",
address = "USA United States",

}

TY - GEN

T1 - Power scalable implementation of artificial neural networks

AU - Modi, S. S.

AU - Wilson, P. R.

AU - Brown, A. D.

PY - 2005

Y1 - 2005

N2 - As the use of Artificial Neural Network(ANN) in mobile embedded devices gets more pervasive, power consumption of ANN hardware is becoming a major limiting factor. Although considerable research efforts are now directed towards low-power implementations of ANN, the issue of dynamic power scalability of the implemented design has been largely overlooked. In this paper, we discuss the motivation and basic principles for implementing power scaling in ANN Hardware. With the help of a simple example, we demonstrate how power scaling can be achieved with dynamic pruning techniques.

AB - As the use of Artificial Neural Network(ANN) in mobile embedded devices gets more pervasive, power consumption of ANN hardware is becoming a major limiting factor. Although considerable research efforts are now directed towards low-power implementations of ANN, the issue of dynamic power scalability of the implemented design has been largely overlooked. In this paper, we discuss the motivation and basic principles for implementing power scaling in ANN Hardware. With the help of a simple example, we demonstrate how power scaling can be achieved with dynamic pruning techniques.

UR - http://dx.doi.org/10.1109/ICECS.2005.4633538

U2 - 10.1109/ICECS.2005.4633538

DO - 10.1109/ICECS.2005.4633538

M3 - Conference contribution

SN - 9789972611001

SP - 1

EP - 4

BT - 12th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2005

PB - IEEE

ER -