Uncertainty propagation in neural networks for sparse coding

Danil Kuzin, Olga Isupova, Lyudmila Mihaylova

Research output: Working paper / PreprintPreprint

13 Downloads (Pure)


A novel method to propagate uncertainty through the soft-thresholding nonlinearity is proposed in this paper. At every layer the current distribution of the target vector is represented as a spike and slab distribution, which represents the probabilities of each variable being zero, or Gaussian-distributed. Using the proposed method of uncertainty propagation, the gradients of the logarithms of normalisation constants are derived, that can be used to update a weight distribution. A novel Bayesian neural network for sparse coding is designed utilising both the proposed method of uncertainty propagation and Bayesian inference algorithm.
Original languageEnglish
Publication statusPublished - 29 Nov 2018

Bibliographical note

Presented at the third workshop on Bayesian Deep Learning (NeurIPS 2018)


  • stat.ML
  • cs.LG


Dive into the research topics of 'Uncertainty propagation in neural networks for sparse coding'. Together they form a unique fingerprint.

Cite this