DP-TOST provides a simulation-calibrated differentially private equivalence testing procedure for means and proportions that controls type-I error and recovers power as privacy budget or sample size grows.
Finite Sample Differentially Private Confidence Intervals
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
A random-projection differentially private kernel ERM method attains minimax-optimal excess risk bounds for squared and Lipschitz-smooth convex losses under local strong convexity, plus the first dimension-free bounds for objective-perturbation private linear ERM.
citing papers explorer
-
Equivalence Testing Under Privacy Constraints
DP-TOST provides a simulation-calibrated differentially private equivalence testing procedure for means and proportions that controls type-I error and recovers power as privacy budget or sample size grows.
-
Optimal differentially private kernel learning with random projection
A random-projection differentially private kernel ERM method attains minimax-optimal excess risk bounds for squared and Lipschitz-smooth convex losses under local strong convexity, plus the first dimension-free bounds for objective-perturbation private linear ERM.