Training generative models from privatized data

arXiv:2306.09547v2 Announce Type: replace-cross
Abstract: Local differential privacy is a powerful method for privacy-preserving data collection. In this paper, we develop a framework for training Generative Adversarial Networks (GANs) on differentially privatized data. We show that entropic regularization of optimal transport - a popular regularization method in the literature that has often been leveraged for its computational benefits - enables the generator to learn the raw (unprivatized) data distribution even though it only has access to privatized samples. We prove …

adversarial arxiv collection cs.cr cs.it cs.lg data data collection differential privacy framework gans generative generative adversarial networks generative models literature local math.it networks popular privacy training transport