Variational Model Inversion Attacks. (arXiv:2201.10787v1 [cs.LG])

Web: http://arxiv.org/abs/2201.10787

Given the ubiquity of deep neural networks, it is important that these models
do not reveal information about sensitive data that they have been trained on.
In model inversion attacks, a malicious user attempts to recover the private
dataset used to train a supervised neural network. A successful model inversion
attack should generate realistic and diverse samples that accurately describe
each of the classes in the private dataset. In this work, we provide a
probabilistic interpretation of model inversion attacks, …

arxiv attacks model