Exploring spiking neural networks (SNN) for low Size, Weight, and Power (SWaP) benefits

Trevor J. Bihl, Patrick Farr, Gaetano Di Caterina, Paul Kirkland, Alex Vicente Sola, Davide Manna, Jundong Liu, Kara Combs

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

23 Downloads (Pure)

Abstract

Size, Weight, and Power (SWaP) concerns are growing as artificial intelligence (AI) use spreads in edge applications. AI algorithms, such as artificial neural networks (ANNs), have revolutionized many fields, e.g. computer vision (CV), but at a large computational/power burden. Biological intelligence is notably more computationally efficient. Neuromorphic edge processors and spiking neural networks (SNNs) aim to follow biology closer with spike-based operations resulting in sparsity and lower-SWaP operations than traditional ANNs with SNNs only “firing/spiking” when needed. Understanding the trade space of SWaP when embracing neuromorphic computing has not been studied heavily. To addresses this, we present a repeatable and scalable apples-to-apples comparison of traditional ANNs and SNNs for edge processing with demonstration on both classical and neuromorphic edge hardware. Results show that SNNs combined with neuromorphic hardware can provide comparable accuracy for CV to ANNs at 1/10th the power.
Original languageEnglish
Title of host publicationProceedings of the 57th Hawaii International Conference on System Sciences
EditorsTung X. Bui
Place of PublicationHonolulu, HI
Pages7561-7570
Number of pages10
Publication statusPublished - 3 Jan 2024
Event57th Annual Hawaii International Conference on System Sciences - Honolulu, United States
Duration: 3 Jan 20246 Mar 2024

Publication series

NameProceedings of the Annual Hawaii International Conference on System Sciences
ISSN (Electronic)2572-6862

Conference

Conference57th Annual Hawaii International Conference on System Sciences
Abbreviated titleHICSS24
Country/TerritoryUnited States
CityHonolulu
Period3/01/246/03/24

Keywords

  • swap
  • neural networks
  • neuromorphics
  • edge pocessing
  • deep learning
  • Intelligent Edge Computing in Pervasive Environments

Fingerprint

Dive into the research topics of 'Exploring spiking neural networks (SNN) for low Size, Weight, and Power (SWaP) benefits'. Together they form a unique fingerprint.

Cite this