Abstract
Sparse histogram methods can be useful for returning differentially private counts of items in large or infinite histograms or large group-by queries, and more generally, releasing a set of statistics with sufficient item counts. We consider the Gaussian version of the sparse histogram mechanism and study the exact ϵ, δ differential privacy guarantees satisfied by this mechanism. We compare these exact ϵ, δ parameters to the simpler overestimates used in prior work to quantify the impact of looser privacy bounds.
| Original language | English (US) |
|---|---|
| Journal | Journal of Privacy and Confidentiality |
| Volume | 14 |
| Issue number | 1 |
| DOIs | |
| State | Published - 2024 |
All Science Journal Classification (ASJC) codes
- Computer Science (miscellaneous)
- Statistics and Probability
- Computer Science Applications