• Corpus ID: 231879873

Uncertainty Propagation in Convolutional Neural Networks: Technical Report

@article{Tzelepis2021UncertaintyPI,
  title={Uncertainty Propagation in Convolutional Neural Networks: Technical Report},
  author={Christos Tzelepis and I. Patras},
  journal={ArXiv},
  year={2021},
  volume={abs/2102.06064}
}
In this technical report we study the problem of propagation of uncertainty (in terms of variances of given univariate normal random variables) through typical building blocks of a Convolutional Neural Network (CNN). These include layers that perform linear operations, such as 2D convolutions, fullyconnected, and average pooling layers, as well as layers that act non-linearly on their input, such as the Rectified Linear Unit (ReLU). Finally, we discuss the sigmoid function, for which we give… 
1 Citations

Figures from this paper

Propagating Variational Model Uncertainty for Bioacoustic Call Label Smoothing

Georgios Rizos, Jenna Lawson, Simon Mitchell, Pranay Shah, Xin Wen, Cristina Banks-Leite, Robert Ewers and Björn W. Schuller 1*GLAM – Group on Language, Audio, & Music, Department of Computing,

References

Semi-analytical approximations to statistical moments of sigmoid and softmax mappings of normal variables

This technical note is concerned with accurate and computationally efficient approximations of moments of Gaussian random variables passed through sigmoid or softmax mappings, which are semi-analytical and highly accurate.