TCL: an ANN-to-SNN Conversion with Trainable Clipping Layers

Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to analog-neural-networks (ANNs). Although it is difficult to efficiently train SNNs, many techniques to convert trained ANNs to SNNs have been developed...

Full description

Saved in:
Bibliographic Details
Published in2021 58th ACM/IEEE Design Automation Conference (DAC) pp. 793 - 798
Main Authors Ho, Nguyen-Dong, Chang, Ik-Joon
Format Conference Proceeding
LanguageEnglish
Published IEEE 05.12.2021
Subjects
ANN-to-SNN conversion
Clocks
Design automation
Limiting
Neural networks
Spiking Neural Network
trainable clipping layer
Training
Online AccessGet full text
DOI10.1109/DAC18074.2021.9586266

Cover

More Information
Summary:Spiking-neural-networks (SNNs) are promising at edge devices since the event-driven operations of SNNs provides significantly lower power compared to analog-neural-networks (ANNs). Although it is difficult to efficiently train SNNs, many techniques to convert trained ANNs to SNNs have been developed. However, after the conversion, a trade-off relation between accuracy and latency exists in SNNs, causing considerable latency in large size datasets such as ImageNet. We present a technique, named as TCL, to alleviate the trade-off problem, enabling the accuracy of 73.87% (VGG-16) and 70.37% (ResNet-34) for ImageNet with the moderate latency of 250 cycles in SNNs.
DOI:10.1109/DAC18074.2021.9586266