SpArNet: Sparse Asynchronous Neural Network execution for energy efficient inference

Mina A. Khoei, Amirreza Yousefzadeh, Arash Pourtaherian, Orlando Moreira, Jonathan Tapson

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

Samenvatting

Biological neurons are known to have sparse and asynchronous communications using spikes. Despite our incomplete understanding of processing strategies of the brain, its low energy consumption in fulfilling delicate tasks suggests the existence of energy efficient mechanisms. Inspired by these key factors, we introduce SpArNet, a bio-inspired quantization scheme to convert a pre-trained convolutional neural network to a spiking neural network, with the aim of minimizing the computational load for execution on neuromorphic processors. The proposed scheme has significant advantages over the reference CNN in a reduced number of synaptic operations, and can be used for frequent executions of inference tasks. The computational load of SpArNet is adjusted to the spatio-temporal dynamics of the the input data. We have tested the converted network on two applications (autonomous steering and hand gesture recognition), demonstrating a significant reduction on the number of required synaptic operations.

Originele taal-2Engels
TitelProceedings - 2020 IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2020
UitgeverijInstitute of Electrical and Electronics Engineers
Pagina's256-260
Aantal pagina's5
ISBN van elektronische versie9781728149226
DOI's
StatusGepubliceerd - aug 2020
Evenement2020 IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2020 - Genova, Italië
Duur: 31 aug 20202 sep 2020

Congres

Congres2020 IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2020
LandItalië
StadGenova
Periode31/08/202/09/20

Vingerafdruk Duik in de onderzoeksthema's van 'SpArNet: Sparse Asynchronous Neural Network execution for energy efficient inference'. Samen vormen ze een unieke vingerafdruk.

  • Citeer dit

    Khoei, M. A., Yousefzadeh, A., Pourtaherian, A., Moreira, O., & Tapson, J. (2020). SpArNet: Sparse Asynchronous Neural Network execution for energy efficient inference. In Proceedings - 2020 IEEE International Conference on Artificial Intelligence Circuits and Systems, AICAS 2020 (blz. 256-260). [9073827] Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/AICAS48895.2020.9073827