Federated Generative Privacy
This addresses privacy concerns for users in federated learning systems, though it appears incremental as it builds on existing GAN and FedAvg methods.
The paper tackles the problem of privacy-preserving data release in federated learning by proposing FedGP, a framework that uses generative adversarial networks trained with FedAvg to generate high-quality labeled artificial data, and demonstrates that it significantly reduces vulnerability to model inversion attacks.
In this paper, we propose FedGP, a framework for privacy-preserving data release in the federated learning setting. We use generative adversarial networks, generator components of which are trained by FedAvg algorithm, to draw privacy-preserving artificial data samples and empirically assess the risk of information disclosure. Our experiments show that FedGP is able to generate labelled data of high quality to successfully train and validate supervised models. Finally, we demonstrate that our approach significantly reduces vulnerability of such models to model inversion attacks.