Optimal Differentially Private Sampling of Unbounded Gaussians
By: Valentio Iverson, Gautam Kamath, Argyris Mouzakis
Potential Business Impact:
Lets computers make private data from random numbers.
We provide the first $\widetilde{\mathcal{O}}\left(d\right)$-sample algorithm for sampling from unbounded Gaussian distributions under the constraint of $\left(\varepsilon, \delta\right)$-differential privacy. This is a quadratic improvement over previous results for the same problem, settling an open question of Ghazi, Hu, Kumar, and Manurangsi.
Similar Papers
Optimizing Noise Distributions for Differential Privacy
Information Theory
Protects private data better while sharing it.
Differential Privacy and Survey Sampling
Statistics Theory
Protects private data when counting people.
An $\tilde{O}$ptimal Differentially Private Learner for Concept Classes with VC Dimension 1
Machine Learning (CS)
Makes learning private and super fast.