PATE-TripleGAN: Privacy-Preserving Image Synthesis with Gaussian Differential Privacy

arXiv:2404.12730v1 Announce Type: cross
Abstract: Conditional Generative Adversarial Networks (CGANs) exhibit significant potential in supervised learning model training by virtue of their ability to generate realistic labeled images. However, numerous studies have indicated the privacy leakage risk in CGANs models. The solution DPCGAN, incorporating the differential privacy framework, faces challenges such as heavy reliance on labeled data for model training and potential disruptions to original gradient information due to excessive gradient clipping, making it difficult to ensure model accuracy. To …

adversarial arxiv challenges cs.cr cs.cv cs.lg differential privacy framework generative generative adversarial networks image images model training networks privacy privacy framework risk solution studies training