A bridge between features and evidence for binary attribute-driven perfect privacy
Paul-Gauthier Noé, Andreas Nautsch, Driss Matrouf, Pierre-Michel Bousquet, Jean-François Bonastre
Attribute-driven privacy aims to conceal a single user's attribute, contrary
to anonymisation that tries to hide the full identity of the user in some data.
When the attribute to protect from malicious inferences is binary, perfect
privacy requires the log-likelihood-ratio to be zero resulting in no
strength-of-evidence. This work presents an approach based on normalizing flow
that maps a feature vector into a latent space where the strength-of-evidence,
related to the binary attribute, and an independent residual are disentangled.
It can be seen as a non-linear discriminant analysis where the mapping is
invertible allowing generation by mapping the latent variable back to the
original space. This framework allows to manipulate the log-likelihood-ratio of
the data and thus to set it to zero for privacy. We show the applicability of
the approach on an attribute-driven privacy task where the sex information is
removed from speaker embeddings. Results on VoxCeleb2 dataset show the
efficiency of the method that outperforms in terms of privacy and utility our
previous experiments based on adversarial disentanglement.