Classification Performances due to Asymmetric Nonlinear Weight Updates in Analog Artificial Synapse-Based Hardware Neural Networks

Yeon Pyo, Sahn Nahm, Jichai Jeong

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

Abstract

Artificial synapses are fundamental for neuromorphic computing to overcome the bottleneck of the von Neumann system. In particular, a memristor synapse-based neuromorphic system has been known as an optimal device for effectively implementing a hardware neural network. Here, we propose the memristor synapse which shows potentiation and depression process like biological brain mechanisms and investigate the effects of varying the device parameters of nonlinearity and asymmetry on the classification accuracy. We find that the virtual devices with a nonlinearity of less than 10 can be obtained the classification accuracy up to 80%. Our approach demonstrates a practical neuromorphic system based on virtual device on simulation and measured device on experiment and verifies the feasibility of the hardware neural networks.

Original languageEnglish
Title of host publication10th International Winter Conference on Brain-Computer Interface, BCI 2022
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665413374
DOIs
Publication statusPublished - 2022
Event10th International Winter Conference on Brain-Computer Interface, BCI 2022 - Gangwon-do, Korea, Republic of
Duration: 2022 Feb 212022 Feb 23

Publication series

NameInternational Winter Conference on Brain-Computer Interface, BCI
Volume2022-February
ISSN (Print)2572-7672

Conference

Conference10th International Winter Conference on Brain-Computer Interface, BCI 2022
Country/TerritoryKorea, Republic of
CityGangwon-do
Period22/2/2122/2/23

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Keywords

  • Artificial synapse
  • Memristor
  • Neural network
  • Neuromorphic system

ASJC Scopus subject areas

  • Artificial Intelligence
  • Human-Computer Interaction
  • Signal Processing

Fingerprint

Dive into the research topics of 'Classification Performances due to Asymmetric Nonlinear Weight Updates in Analog Artificial Synapse-Based Hardware Neural Networks'. Together they form a unique fingerprint.

Cite this