Variation-aware binarized memristive networks

Corey Lammie, Olga Krestinskaya, Alex James, Mostafa Rahimi Azghadi

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

1 Citation (Scopus)

Abstract

The quantization of weights to binary states in Deep Neural Networks (DNNs) can replace resource-hungry multiply accumulate operations with simple accumulations. Such Binarized Neural Networks (BNNs) exhibit greatly reduced resource and power requirements. In addition, memristors have been shown as promising synaptic weight elements in DNNs. In this paper, we propose and simulate novel Binarized Memristive Convolutional Neural Network (BMCNN) architectures employing hybrid weight and parameter representations. We train the proposed architectures offline and then map the trained parameters to our binarized memristive devices for inference. To take into account the variations in memristive devices, and to study their effect on the performance, we introduce variations in RON and ROFF. Moreover, we introduce means to mitigate the adverse effect of memristive variations in our proposed networks. Finally, we benchmark our BMCNNs and variation-aware BMCNNs using the MNIST dataset.

Original languageEnglish
Title of host publication2019 26th IEEE International Conference on Electronics, Circuits and Systems, ICECS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages490-493
Number of pages4
ISBN (Electronic)9781728109961
DOIs
Publication statusPublished - Nov 2019
Event26th IEEE International Conference on Electronics, Circuits and Systems, ICECS 2019 - Genoa, Italy
Duration: Nov 27 2019Nov 29 2019

Publication series

Name2019 26th IEEE International Conference on Electronics, Circuits and Systems, ICECS 2019

Conference

Conference26th IEEE International Conference on Electronics, Circuits and Systems, ICECS 2019
CountryItaly
CityGenoa
Period11/27/1911/29/19

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Control and Optimization
  • Computer Networks and Communications
  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Variation-aware binarized memristive networks'. Together they form a unique fingerprint.

Cite this