Fixed-Point Implementation of Convolutional Neural Networks for Image Classification

Chun Y. Lo, Francis C.M. Lau, Chiu Wing Sham

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

24 Citations (Scopus)

Abstract

In this paper, we show step-by-step how to design and optimize a fixed-point convolutional neural network (CNN) classifier. Moreover, the fixed-point classifier has been implemented using C++ programming and onto an FPGA. We show that the classifier with 4-bit fixed-point arithmetic with 8bit additions can classify handwritten digits with over 99.16% accuracy. Compared with the floating-point classifier which achieves 99.55% accuracy, the degradation due to fixed-point implementation is less than 0.4%.

Original languageEnglish
Title of host publicationProceedings of 2018 International Conference on Advanced Technologies for Communications, ATC 2018
EditorsVo Nguyen Quoc Bao, Tran Trung Duy
PublisherIEEE Computer Society
Pages105-109
Number of pages5
ISBN (Electronic)9781538661130
DOIs
Publication statusPublished - 18 Oct 2018
Event11th International Conference on Advanced Technologies for Communications, ATC 2018 - Ho Chi Minh City, Viet Nam
Duration: 18 Oct 201820 Oct 2018

Publication series

NameInternational Conference on Advanced Technologies for Communications
Volume2018-October
ISSN (Print)2162-1039
ISSN (Electronic)2162-1020

Conference

Conference11th International Conference on Advanced Technologies for Communications, ATC 2018
Country/TerritoryViet Nam
CityHo Chi Minh City
Period18/10/1820/10/18

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Hardware and Architecture
  • Software

Fingerprint

Dive into the research topics of 'Fixed-Point Implementation of Convolutional Neural Networks for Image Classification'. Together they form a unique fingerprint.

Cite this