Privacy Amplification for the Gaussian Mechanism via Bounded Support

arXiv:2403.05598v1 Announce Type: new
Abstract: Data-dependent privacy accounting frameworks such as per-instance differential privacy (pDP) and Fisher information loss (FIL) confer fine-grained privacy guarantees for individuals in a fixed training dataset. These guarantees can be desirable compared to vanilla DP in real world settings as they tightly upper-bound the privacy leakage for a $\textit{specific}$ individual in an $\textit{actual}$ dataset, rather than considering worst-case datasets. While these frameworks are beginning to gain popularity, to date, there is a lack of private …

accounting amplification arxiv can cs.cr cs.lg data dataset differential privacy frameworks information instance loss mechanism privacy real settings support training vanilla world