Learning Semi-Supervised Anonymized Representations by Mutual Information

Clément Feutry, Pablo Piantanida, Pierre DuhamelICASSP 2020 · OpenReview

Abstract

This paper addresses the problem of removing from a set of data (here images) a given private information, while still allowing other utilities on the processed data. This is obtained by training concurrently a GAN-like discriminator and an autoencoder. The optimization of the resulting structure involves a novel surrogate of the misclassification probability of the information to remove. Several examples are given, demonstrating that a good level of privacy can be obtained on images at the cost of the introduction of very small artifacts.
BibTeX
@inproceedings{feutry2020learning,
  author = {Clément Feutry and Pablo Piantanida and Pierre Duhamel},
  title = {{Learning Semi-Supervised Anonymized Representations by Mutual Information}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2020},
  url = {https://openreview.net/forum?id=fBj0VRZOaLp},
}

Loading related content...

Figure