Shake-Shake Regularization of 3-Branch Residual Networks

Abstract

The method introduced in this paper aims at helping computer vision practitioners faced with an overfit problem. The idea is to replace, in a 3-branch ResNet, the standard summation of residual branches by a stochastic affine combination. The largest tested model improves on the best single shot published result on CIFAR-10 by reaching 2.86% test error. Code is available at https://github.com/xgastaldi/shake-shake

Cite

Text

Gastaldi. "Shake-Shake Regularization of 3-Branch Residual Networks." International Conference on Learning Representations, 2017.

Markdown

[Gastaldi. "Shake-Shake Regularization of 3-Branch Residual Networks." International Conference on Learning Representations, 2017.](https://mlanthology.org/iclr/2017/gastaldi2017iclr-shake/)

BibTeX

@inproceedings{gastaldi2017iclr-shake,
  title     = {{Shake-Shake Regularization of 3-Branch Residual Networks}},
  author    = {Gastaldi, Xavier},
  booktitle = {International Conference on Learning Representations},
  year      = {2017},
  url       = {https://mlanthology.org/iclr/2017/gastaldi2017iclr-shake/}
}