Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners

Rachel Emily Redberg, Antti Koskela, Yu-Xiang Wang

Keywords: differential privacy, empirical risk minimization, objective perturbation

TL;DR: We provide the old-school objective perturbation mechanism with new privacy analyses and computational tools.

Abstract: In the arena of privacy-preserving machine learning, differentially private stochastic gradient descent (DP-SGD) has outstripped the objective perturbation mechanism in popularity and interest. Though unrivaled in versatility, DP-SGD requires a non-trivial privacy overhead (for privately tuning the model’s hyperparameters) and a computational complexity which might be extravagant for simple models such as linear and logistic regression. This paper revamps the objective perturbation mechanism with tighter privacy analyses and new computational tools that boost it to perform competitively with DP-SGD on unconstrained convex generalized linear problems.

Submission Number: 9090