Metric Differential Privacy at the User-Level

arXiv:2405.02665v1 Announce Type: new
Abstract: Metric differential privacy (DP) provides heterogeneous privacy guarantees based on a distance between the pair of inputs. It is a widely popular notion of privacy since it captures the natural privacy semantics for many applications (such as, for location data) and results in better utility than standard DP. However, prior work in metric DP has primarily focused on the \textit{item-level} setting where every user only reports a single data item. A more realistic setting is …

applications arxiv cs.cr data differential privacy inputs location location data metric natural notion popular privacy results semantics standard utility