Differential Privacy in Two-Layer Networks: How DP-SGD Harms Fairness and Robustness
Differentially private learning is essential for training models on sensitive data, but empirical studies consistently show that it can degrade performance, introduce fairness issues like disparate impact, and reduce adversarial robustness. The theoretical underpinnings of these phenomena in modern, non-convex neural networks remain largely unexplored. This paper introduces a unified feature-centric framework to analyze the feature learning dynamics of differentially private stochastic gradient descent (DP-SGD) in two-layer ReLU convolutional neural networks. Our analysis establishes test loss bounds governed by a crucial metric: the feature-to-noise ratio (FNR). We demonstrate that the noise required for privacy leads to suboptimal feature learning, and specifically show that: 1) imbalanced FNRs across classes and subpopulations cause disparate impact; 2) even in the same class, noise has a greater negative impact on semantically long-tailed data; and 3) noise injection exacerbates vulnerability to adversarial attacks. Furthermore, our analysis reveals that the popular paradigm of public pre-training and private fine-tuning does not guarantee improvement, particularly under significant feature distribution shifts between datasets. Experiments on synthetic and real-world data corroborate our theoretical findings.
Code (0)
등록된 구현이 없습니다.
Tasks
Adversarial RobustnessSimilar Papers 제목 키워드 기반
RESFL: An Uncertainty-Aware Framework for Responsible Federated Learning by Balancing Privacy, Fairness and Utility in Autonomous Vehicles
Autonomous vehicles (AVs) increasingly rely on Federated Learning (FL) to enhance perception models while preserving privacy. However, existing FL frameworks struggle to balance privacy, fairness, and robustness, leading…
Autonomous VehiclesDisentanglementFairnessFederated Learning+3Towards Explainable Federated Learning: Understanding the Impact of Differential Privacy
Data privacy and eXplainable Artificial Intelligence (XAI) are two important aspects for modern Machine Learning systems. To enhance data privacy, recent machine learning models have been designed as a Federated Learning…
Federated LearningFairness Meets Privacy: Integrating Differential Privacy and Demographic Parity in Multi-class Classification
The increasing use of machine learning in sensitive applications demands algorithms that simultaneously preserve data privacy and ensure fairness across potentially sensitive sub-populations. While privacy and fairness h…
Multi-class ClassificationTowards Trustworthy Federated Learning
This paper develops a comprehensive framework to address three critical trustworthy challenges in federated learning (FL): robustness against Byzantine attacks, fairness, and privacy preservation. To improve the system's…
FairnessFederated LearningFair Differentially Private Federated Learning Framework
Federated learning (FL) is a distributed machine learning strategy that enables participants to collaborate and train a shared model without sharing their individual datasets. Privacy and fairness are crucial considerati…
FairnessFederated Learning