Differential Privacy (DP) image data synthesis, which leverages the DP technique to generate synthetic data to replace the sensitive data, allowing organizations to share and …
S Lyu, MF Liu, M Vinaroz, M Park - arXiv preprint arXiv:2305.15759, 2023 - arxiv.org
Diffusion models (DMs) are widely used for generating high-quality high-dimensional images in a non-differentially private manner. To address this challenge, recent papers …
Maximum mean discrepancy (MMD) is a particularly useful distance metric for differentially private data generation: when used with finite-dimensional features it allows us to …
Differential Privacy (DP) image data synthesis, which leverages the DP technique to generate synthetic data to replace the sensitive data, allowing organizations to share and …
H Wang, S Pang, Z Lu, Y Rao, Y Zhou, M Xue - 2024 - usenix.org
Utilizing sensitive images (eg, human faces) for training DL models raises privacy concerns. One straightforward solution is to replace the private images with synthetic ones generated …
Safeguarding privacy in sensitive training data is paramount, particularly in the context of generative modeling. This is done through either differentially private stochastic gradient …
Text-to-image diffusion models have been shown to suffer from sample-level memorization, possibly reproducing near-perfect replica of images that they are trained on, which may be …
Differentially private synthetic data is a promising alternative for sensitive data release. Many differentially private generative models have been proposed in the literature. Unfortunately …
H Sha, Y Cao, Y Liu, Y Wu, R Liu, H Chen - arXiv preprint arXiv …, 2024 - arxiv.org
Differentially Private Stochastic Gradient Descent (DPSGD) is widely utilized to preserve training data privacy in deep learning, which first clips the gradients to a predefined norm …