Work

Improved Rates of Differentially Private Nonconvex-Strongly-Concave Minimax Optimization

Year: 2025

Type: conference-paper

Abstract: In this paper, we study the problem of (finite sum) minimax optimization in the Differential Privacy (DP) model. Unlike most of the previous studies on the (strongly) convex-concave settings or loss functions satisfying the Polyak-Lojasiewicz condition, here we mainly focus on the nonconvex-strongly-concave one, which encapsulates many models in deep learning such as deep AUC maximization. Specifically, we first analyze a DP version of Stochastic Gradient Descent Ascent (SGDA) and show the utility bound in terms of the Euclidean norm of the gradient for the empirical risk function. We then propose a new method with less gradient noise variance and improve the upper bound to the best-known result for DP Empirical Risk Minimization with non-convex loss. We also discussed several lower bounds of private minimax optimization. Finally, experiments on AUC maximization, generative adversarial networks, and temporal difference learning with real-world data support our theoretical analysis.

Source: Proceedings of the AAAI Conference on Artificial Intelligence

DOI: 10.1609/aaai.v39i21.34410

Authors: Ruijia Zhang, Mingxi Lei, Meng Ding, Zihang Xiang, Jinhui Xu, Di Wang

Institutions: Johns Hopkins University, Chinese University of Hong Kong, Shenzhen, University at Buffalo, State University of New York, King Abdullah University of Science and Technology

Language: English

Topic: Privacy-Preserving Technologies in Data

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: diamond

Funder: King Abdullah University of Science and Technology

Awards: G3214865012, G3518144781, G419305456, G4795079661