Evolving Ensembles of Spiking Neural Networks for Neuromorphic Systems

Daniel Elbrecht, Shruti R. Kulkarni, Maryam Parsa, J. Parker Mitchell, Catherine D. Schuman

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Evolutionary algorithms have been proposed as a solution to overcome many of the challenges associated with training spiking neural networks. While evolutionary optimization for spiking neural networks is very flexible, its performance has difficulty scaling to complex tasks and correspondingly complex network structures. Here we propose a method for evolving ensembles of spiking neural networks. By using ensemble learning, the flexibility of evolutionary optimization is fully preserved while scaling to more challenging tasks. We test the performance of the proposed method using handwritten digit classification. We investigate multiple strategies for constructing ensembles of spiking neural networks, and demonstrate that evolving ensembles of SNNs offers significant performance advantages over evolutionary optimization.

Original languageEnglish
Title of host publication2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1989-1994
Number of pages6
ISBN (Electronic)9781728125473
DOIs
StatePublished - Dec 1 2020
Externally publishedYes
Event2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020 - Virtual, Canberra, Australia
Duration: Dec 1 2020Dec 4 2020

Publication series

Name2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020

Conference

Conference2020 IEEE Symposium Series on Computational Intelligence, SSCI 2020
Country/TerritoryAustralia
CityVirtual, Canberra
Period12/1/2012/4/20

Keywords

  • ensembles
  • evolutionary algorithms
  • neuromorphic computing
  • spiking neural networks

Fingerprint

Dive into the research topics of 'Evolving Ensembles of Spiking Neural Networks for Neuromorphic Systems'. Together they form a unique fingerprint.

Cite this