Minimizing Inference Time: Optimization Methods for Converted Deep Spiking Neural Networks

Etienne Mueller, Julius Hansjakob, Daniel Auge, Alois Knoll

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

5 Scopus citations

Abstract

Spiking neural networks offer the potential to drastically reduce energy consumption in edge devices. Unfortunately they are overshadowed by today's common analog neural networks, whose superior backpropagation-based learning algorithms frequently demonstrate superhuman performance on different tasks. The best accuracies in spiking networks are achieved by training analog networks and converting them. Still, during runtime many simulation time steps are needed until they converge. To improve the simulation time we evaluate two inference optimization algorithms and propose an additional method for error minimization. We assess them on Residual Networks of different sizes, up to ResNet101. The combination of all three is evaluated on a large scale with a RetinaNet on the COCO dataset. Our experiments show that all optimization algorithms combined can speed up the inference process by a factor of ten. Additionally, the accuracy loss between the original and the converted network is less than half a percent, which is the lowest on a complex dataset reported to date.

Original languageEnglish
Title of host publicationIJCNN 2021 - International Joint Conference on Neural Networks, Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780738133669
DOIs
StatePublished - 18 Jul 2021
Event2021 International Joint Conference on Neural Networks, IJCNN 2021 - Virtual, Shenzhen, China
Duration: 18 Jul 202122 Jul 2021

Publication series

NameProceedings of the International Joint Conference on Neural Networks
Volume2021-July

Conference

Conference2021 International Joint Conference on Neural Networks, IJCNN 2021
Country/TerritoryChina
CityVirtual, Shenzhen
Period18/07/2122/07/21

Keywords

  • conversion
  • neuromorphic computing
  • object detection
  • residual networks
  • spiking neural networks

Fingerprint

Dive into the research topics of 'Minimizing Inference Time: Optimization Methods for Converted Deep Spiking Neural Networks'. Together they form a unique fingerprint.

Cite this