PAC Privacy Preserving Diffusion Models
Abstract: Data privacy protection is garnering increased attention among researchers. Diffusion models (DMs), particularly with strict differential privacy, can potentially produce images with both high privacy and visual quality. However, challenges arise such as in ensuring robust protection in privatizing specific data attributes, areas where current models often fall short. To address these challenges, we introduce the PAC Privacy Preserving Diffusion Model, a model leverages diffusion principles and ensure Probably Approximately Correct (PAC) privacy. We enhance privacy protection by integrating a private classifier guidance into the Langevin Sampling Process. Additionally, recognizing the gap in measuring the privacy of models, we have developed a novel metric to gauge privacy levels. Our model, assessed with this new metric and supported by Gaussian matrix computations for the PAC bound, has shown superior performance in privacy protection over existing leading private generative models according to benchmark tests.
Submission Length: Regular submission (no more than 12 pages of main content)
Changes Since Last Submission: We've thoroughly reviewed the unclear statement of PAC privacy in Section 3.2 and unexplained symbols concept you mentioned in Section 3.2 and 3.3. Besides, other revisions are also included in the paper. We are committed to making the necessary revisions to improve the quality of our paper.
Assigned Action Editor: ~Antti_Koskela1
Submission Number: 2527
Loading