{"id":"arxiv/2608.31138","url":"https://arxiv.org/abs/2608.31138","tags":["math.ST","stat.ME"],"score":0,"title":"Scale Analysis and Shape Selection for the Generalized Gaussian Mechanism under Approximate Differential Privacy","domain":"math.ST","source":"arxiv","authors":["Xiang Zhang","Mohamedou Ould Haye","Yiqiang Q. Zhao"],"reviews":0,"subject":"math.ST","summary":"Differential privacy provides a rigorous framework for protecting private information, typically achieved by adding random noise to query results. The generalized Gaussian family is a flexible class of additive noise distributions indexed by the shape parameter $p$ and includes the Laplace and Gaussian distributions as special cases $p=1$ and $p=2$, respectively. This paper studies the privacy-feasible scale estimation and the shape parameter selection of the generalized Gaussian mechanism (GGM) under $(\\varepsilon,δ)$-differential privacy. For a given sensitivity vector $Δ$ and…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-08-31","reactions":{}}