An energy-efficient memory-based high-throughput VLSI architecture for convolutional networks

Mingu Kang, Sujan K. Gonugondla, Min Sun Keel, Naresh R. Shanbhag

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

Abstract

In this paper, an energy efficient, memory-intensive, and high throughput VLSI architecture is proposed for convolutional networks (C-Net) by employing compute memory (CM) [1], where computation is deeply embedded into the memory (SRAM). Behavioral models incorporating CM's circuit non-idealities and energy models in 45nm SOI CMOS are presented. System-level simulations using these models demonstrate that the probability of handwritten digit recognition Pr > 0.99 can be achieved using the MNIST database [2], along with a 24.5× reduced energy delay product, a 5.0× reduced energy, and a 4.9× higher throughput as compared to the conventional system.

Original languageEnglish (US)
Title of host publication2015 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1037-1041
Number of pages5
ISBN (Electronic)9781467369978
DOIs
StatePublished - Aug 4 2015
Event40th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015 - Brisbane, Australia
Duration: Apr 19 2014Apr 24 2014

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume2015-August
ISSN (Print)1520-6149

Other

Other40th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2015
Country/TerritoryAustralia
CityBrisbane
Period4/19/144/24/14

Keywords

  • Compute memory
  • Convolutional networks
  • Machine learning
  • Pattern recognition

ASJC Scopus subject areas

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'An energy-efficient memory-based high-throughput VLSI architecture for convolutional networks'. Together they form a unique fingerprint.

Cite this