English

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

Computer Vision and Pattern Recognition 2021-12-02 v3

Abstract

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.

Keywords

Cite

@article{arxiv.2008.04509,
  title  = {TCL: an ANN-to-SNN Conversion with Trainable Clipping Layers},
  author = {Nguyen-Dong Ho and Ik-Joon Chang},
  journal= {arXiv preprint arXiv:2008.04509},
  year   = {2021}
}

Comments

To appear in the 58th Design Automation Conference (DAC 2021)

R2 v1 2026-06-23T17:46:08.633Z