BD-NET: A multiplication-less DNN with binarized depthwise separable convolution

Zhezhi He, Shaahin Angizi, Adnan Siraj Rakin, Deliang Fan

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

2 Scopus citations

Abstract

In this work, we propose a multiplication-less deep convolution neural network, called BD-NET. As far as we know, BD-NET is the first to use binarized depthwise separable convolution block as the drop-in replacement of conventional spatial-convolution in deep convolution neural network (CNN). In BD-NET, the computation-expensive convolution operations (i.e. Multiplication and Accumulation) are converted into hardware-friendly Addition/Subtraction operations. In this work, we first investigate and analyze the performance of BD-NET in terms of accuracy, parameter size and computation cost, w.r.t various network configurations. Then, the experiment results show that our proposed BD-NET with binarized depthwise separable convolution can achieve even higher inference accuracy to its baseline CNN counterpart with full-precision conventional convolution layer on the CIFAR-10 dataset. From the perspective of hardware implementation, the convolution layer of BD-NET achieves up to 97.2%, 88.9%, and 99.4% reduction in terms of computation energy, memory usage, and chip area respectively.

Original languageEnglish (US)
Title of host publicationProceedings - 2018 IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2018
PublisherIEEE Computer Society
Pages130-135
Number of pages6
ISBN (Print)9781538670996
DOIs
StatePublished - Aug 7 2018
Event17th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2018 - Hong Kong, Hong Kong
Duration: Jul 9 2018Jul 11 2018

Publication series

NameProceedings of IEEE Computer Society Annual Symposium on VLSI, ISVLSI
Volume2018-July
ISSN (Print)2159-3469
ISSN (Electronic)2159-3477

Conference

Conference17th IEEE Computer Society Annual Symposium on VLSI, ISVLSI 2018
CountryHong Kong
CityHong Kong
Period7/9/187/11/18

Keywords

  • Binarized neural network
  • Multiplication less

ASJC Scopus subject areas

  • Hardware and Architecture
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'BD-NET: A multiplication-less DNN with binarized depthwise separable convolution'. Together they form a unique fingerprint.

Cite this