MLAI 2019 #05. Pablo Piantanida Learning Anonymized Representations with Mutual Information

Speaker: Pablo Piantanida, Associate Professor of Information Theory at CentraleSupelec

Abstract: Statistical methods protecting sensitive information or the identity of the data owner have become critical to ensure privacy of individuals as well as of organizations. In this talk, we present a statistical anonymization method based on representation learning and deep neural networks. Our approach employs adversarial networks to perform a novel variational approximation of the mutual information between the representations and the users identity. We introduce a training objective for simultaneously learning representations that preserve the information of interest (e.g., about regular labels) while dismissing information about the identity of a person (e.g., about private labels). We demonstrate the success of this approach for standard classification versus anonymization tasks.

See others videos of the MLAI workshop on www.mlai-workshop.org