SAU: Smooth Activation Function Using Convolution with Approximate Identities

Abstract

Well-known activation functions like ReLU or Leaky ReLU are non-differentiable at the origin. Over the years, many smooth approximations of ReLU have been proposed using various smoothing techniques. We propose new smooth approximations of a non-differentiable activation function by convolving it with approximate identities. In particular, we present smooth approximations of Leaky ReLU and show that they outperform several well-known activation functions in various datasets and models. We call this function Smooth Activation Unit (SAU). Replacing ReLU by SAU, we get 5.63%, 2.95%, and 2.50% improvement with ShuffleNet V2 (2.0x), PreActResNet 50 and ResNet 50 models respectively on the CIFAR100 dataset and 2.31% improvement with ShuffleNet V2 (1.0x) model on ImageNet-1k dataset.

Cite

Text

Biswas et al. "SAU: Smooth Activation Function Using Convolution with Approximate Identities." Proceedings of the European Conference on Computer Vision (ECCV), 2022. doi:10.1007/978-3-031-19803-8_19

Markdown

[Biswas et al. "SAU: Smooth Activation Function Using Convolution with Approximate Identities." Proceedings of the European Conference on Computer Vision (ECCV), 2022.](https://mlanthology.org/eccv/2022/biswas2022eccv-sau/) doi:10.1007/978-3-031-19803-8_19

BibTeX

@inproceedings{biswas2022eccv-sau,
  title     = {{SAU: Smooth Activation Function Using Convolution with Approximate Identities}},
  author    = {Biswas, Koushik and Kumar, Sandeep and Banerjee, Shilpak and Pandey, Ashish Kumar},
  booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
  year      = {2022},
  doi       = {10.1007/978-3-031-19803-8_19},
  url       = {https://mlanthology.org/eccv/2022/biswas2022eccv-sau/}
}