Differentially Private Algorithms for Smooth Nonconvex ERM

Changyu Gao, Stephen Wright

Keywords: differential privacy, optimization, machine learning, ERM

TL;DR: We develop simple differentially private optimization algorithms that move along directions of (expected) descent to find approximate second-order necessary solutions for non-convex ERM problems.

Abstract: We develop simple differentially private optimization algorithms that move along directions of (expected) descent to find approximate second-order necessary solution for non-convex ERM problems. 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.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Submission Guidelines: Yes

Please Choose The Closest Area That Your Submission Falls Into: Social Aspects of Machine Learning (eg, AI safety, fairness, privacy, interpretability, human-AI interaction, ethics)