PredictiveNet: An energy-efficient convolutional neural network via zero prediction

Yingyan Lin, Charbel Sakr, Yongjune Kim, Naresh Shanbhag

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

Abstract

Convolutional neural networks (CNNs) have gained considerable interest due to their record-breaking performance in many recognition tasks. However, the computational complexity of CNNs precludes their deployments on power-constrained embedded platforms. In this paper, we propose predictive CNN (PredictiveNet), which predicts the sparse outputs of the non-linear layers thereby bypassing a majority of computations. PredictiveNet skips a large fraction of convolutions in CNNs at runtime without modifying the CNN structure or requiring additional branch networks. Analysis supported by simulations is provided to justify the proposed technique in terms of its capability to preserve the mean square error (MSE) of the nonlinear layer outputs. When applied to a CNN for handwritten digit recognition, simulation results show that PredictiveNet can reduce the computational cost by a factor of 2.9χ compared to a state-of-the-art CNN, while incurring marginal accuracy degradation.

Original languageEnglish (US)
Title of host publicationIEEE International Symposium on Circuits and Systems
Subtitle of host publicationFrom Dreams to Innovation, ISCAS 2017 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781467368520
DOIs
StatePublished - Sep 25 2017
Event50th IEEE International Symposium on Circuits and Systems, ISCAS 2017 - Baltimore, United States
Duration: May 28 2017May 31 2017

Publication series

NameProceedings - IEEE International Symposium on Circuits and Systems
ISSN (Print)0271-4310

Other

Other50th IEEE International Symposium on Circuits and Systems, ISCAS 2017
CountryUnited States
CityBaltimore
Period5/28/175/31/17

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'PredictiveNet: An energy-efficient convolutional neural network via zero prediction'. Together they form a unique fingerprint.

Cite this