Federated Learning With Individualized Privacy Through Client Sampling

2025-01-29

AI Summary

To address the privacy-utility trade-off arising from heterogeneous client privacy preferences and non-IID data in federated learning (FL), this paper proposes an adaptive FL framework grounded in individualized differential privacy (IDP). Methodologically, it introduces the SAMPLE sampling mechanism—first applied to FL—to dynamically adjust each client’s local participation probability according to its personalized privacy budget. We further design IDP-FedAvg, which jointly optimizes client sampling rates and explicitly models data heterogeneity. Extensive experiments on multiple benchmark datasets demonstrate that our approach significantly outperforms uniform-DP baselines and SCALE: it achieves superior model convergence and generalization while preserving client-specific privacy guarantees. Notably, it exhibits enhanced robustness under strong non-IID conditions. By enabling fine-grained, client-tailored privacy control without compromising utility, this work establishes a novel paradigm for decentralized privacy-preserving machine learning.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Multiagent Systems: Mechanism Design
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Security and Privacy: Privacy-enhancing technologies
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

With growing concerns about user data collection, individualized privacy has emerged as a promising solution to balance protection and utility by accounting for diverse user privacy preferences. Instead of enforcing a uniform level of anonymization for all users, this approach allows individuals to choose privacy settings that align with their comfort levels. Building on this idea, we propose an adapted method for enabling Individualized Differential Privacy (IDP) in Federated Learning (FL) by handling clients according to their personal privacy preferences. By extending the SAMPLE algorithm from centralized settings to FL, we calculate client-specific sampling rates based on their heterogeneous privacy budgets and integrate them into a modified IDP-FedAvg algorithm. We test this method under realistic privacy distributions and multiple datasets. The experimental results demonstrate that our approach achieves clear improvements over uniform DP baselines, reducing the trade-off between privacy and utility. Compared to the alternative SCALE method in related work, which assigns differing noise scales to clients, our method performs notably better. However, challenges remain for complex tasks with non-i.i.d. data, primarily stemming from the constraints of the decentralized setting.

Problem

• Federated Learning
• Differential Privacy
• Personalized Privacy Settings

Innovation

• Personalized Differential Privacy
• IDP-FedAvg
• Privacy-Preserving Machine Learning