Stable ResNet
Abstract
Deep ResNet architectures have achieved state of the art performance on many tasks. While they solve the problem of gradient vanishing, they might suffer from gradient exploding as the depth becomes large (Yang et al. 2017). Moreover, recent results have shown that ResNet might lose expressivity as the depth goes to infinity (Yang et al. 2017, Hayou et al. 2019). To resolve these issues, we introduce a new class of ResNet architectures, calledStable ResNet, that have the property of stabilizing the gradient while ensuring expressivity in the infinite depth limit.
Cite
Text
Hayou et al. "Stable ResNet." Artificial Intelligence and Statistics, 2021.Markdown
[Hayou et al. "Stable ResNet." Artificial Intelligence and Statistics, 2021.](https://mlanthology.org/aistats/2021/hayou2021aistats-stable/)BibTeX
@inproceedings{hayou2021aistats-stable,
title = {{Stable ResNet}},
author = {Hayou, Soufiane and Clerico, Eugenio and He, Bobby and Deligiannidis, George and Doucet, Arnaud and Rousseau, Judith},
booktitle = {Artificial Intelligence and Statistics},
year = {2021},
pages = {1324-1332},
volume = {130},
url = {https://mlanthology.org/aistats/2021/hayou2021aistats-stable/}
}