BriefGPT.xyz
Nov, 2020
广义线性模型中的隐私成本：算法与极小极大下界
The Cost of Privacy in Generalized Linear Models: Algorithms and Minimax Lower Bounds
HTML
PDF
T. Tony Cai, Yichen Wang, Linjun Zhang
TL;DR
使用构建的差分隐私版本的梯度下降算法，针对低维和高维稀疏广义线性模型提出参数估计，通过表征统计学性能和建立GLMs的隐私约束极小值下界来显示所提算法的近乎速率最优性。
Abstract
The trade-off between differential privacy and statistical accuracy in generalized linear models (GLMs) is studied. We propose
differentially private algorithms
for
parameter estimation
in both low-dimensional an
→