pith. sign in

arxiv: 1703.09947 · v2 · pith:SOUKRNHVnew · submitted 2017-03-29 · 💻 cs.LG · cs.DS· stat.ML

Efficient Private ERM for Smooth Objectives

classification 💻 cs.LG cs.DSstat.ML
keywords privateobjectivessmoothutilityalgorithmalgorithmsdescentefficient
0
0 comments X
read the original abstract

In this paper, we consider efficient differentially private empirical risk minimization from the viewpoint of optimization algorithms. For strongly convex and smooth objectives, we prove that gradient descent with output perturbation not only achieves nearly optimal utility, but also significantly improves the running time of previous state-of-the-art private optimization algorithms, for both $\epsilon$-DP and $(\epsilon, \delta)$-DP. For non-convex but smooth objectives, we propose an RRPSGD (Random Round Private Stochastic Gradient Descent) algorithm, which provably converges to a stationary point with privacy guarantee. Besides the expected utility bounds, we also provide guarantees in high probability form. Experiments demonstrate that our algorithm consistently outperforms existing method in both utility and running time.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Integrating Feature Correlation in Differential Privacy with Applications in DP-ERM

    cs.LG 2026-05 unverdicted novelty 7.0

    CorrDP relaxes standard differential privacy by incorporating feature correlations, enabling distance-dependent noise in DP-ERM for better privacy-utility tradeoffs.

  2. DP-SelFT: Differentially Private Selective Fine-Tuning for Large Language Models

    cs.LG 2026-05 unverdicted novelty 6.0

    DP-SelFT improves the privacy-utility trade-off for LLM fine-tuning by selecting robust layer subsets via DP synthetic data and perturbation-matched evaluation.