Paper ID: 2302.04972
Differentially Private Optimization for Smooth Nonconvex ERM
Changyu Gao, Stephen J. Wright
We develop simple differentially private optimization algorithms that move along directions of (expected) descent to find an approximate second-order solution for nonconvex ERM. We use line search, mini-batching, and a two-phase strategy to improve the speed and practicality of the algorithm. Numerical experiments demonstrate the effectiveness of these approaches.
Submitted: Feb 9, 2023