An efficient synaptic architecture for artificial neural networks

Irem Boybat, Manuel Le Gallo, S. R. Nandakumar, Timoleon Moraitis, Tomas Tuma, Bipin Rajendran, Yusuf Leblebici, Abu Sebastian, Evangelos Eleftheriou

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

1 Scopus citations

Abstract

Artificial neural networks (ANN) have revolutionized the field of machine learning by providing impressive human-like performance in solving real-world tasks in computer vision, speech recognition, or complex strategic games. There is a significant interest in developing non-von Neumann coprocessors for the training of ANNs, where resistive memory devices serve as synaptic elements. However, interdevice variability, limited dynamic range and resolution, nonlinearity and asymmetric switching characteristics pose important technical challenges. We investigate the use of multi-memristive synapses to overcome these challenges. We present a detailed experimental characterization of conductance changes using a phase-change memory chip fabricated in the 90nm technology node and show how multi-memrisive synapses can address the limitations of memristive devices for synaptic implementations. Simulations show that an ANN trained with backpropagation can achieve competitive classification accuracies using such a scheme.

Original languageEnglish (US)
Title of host publication2017 17th Non-Volatile Memory Technology Symposium, NVMTS 2017 - Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1-4
Number of pages4
ISBN (Electronic)9781538604779
DOIs
StatePublished - Dec 8 2017
Event17th Non-Volatile Memory Technology Symposium, NVMTS 2017 - Aachen, Germany
Duration: Aug 30 2017Sep 1 2017

Publication series

Name2017 17th Non-Volatile Memory Technology Symposium, NVMTS 2017 - Conference Proceedings
Volume2017-December

Other

Other17th Non-Volatile Memory Technology Symposium, NVMTS 2017
CountryGermany
CityAachen
Period8/30/179/1/17

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Electronic, Optical and Magnetic Materials

Fingerprint Dive into the research topics of 'An efficient synaptic architecture for artificial neural networks'. Together they form a unique fingerprint.

Cite this