Differentially-Private Decentralized Learning in Heterogeneous Multicast Networks

2025-09-25

AI Summary

This work addresses decentralized learning in heterogeneous multicast networks under differential privacy and node energy budget constraints. We propose a privacy-preserving algorithm that jointly optimizes transmit power control and Gaussian noise injection, tailored to asymmetric channel conditions modeled by row-stochastic adjacency matrices—marking the first such co-design of transmission power and privacy noise levels. Theoretically, we establish an $O(log T)$ convergence rate. Experimentally, under identical privacy budgets ($varepsilon$) and energy constraints, our method achieves significantly higher model accuracy than existing decentralized differentially private approaches, while simultaneously improving communication efficiency, privacy guarantees, and convergence performance.

Technology Category

• Machine Learning: Privacy
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Search and Optimization: Learning to Search

Application Category

• Security and Privacy: Privacy-enhancing technologies
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

We propose a power-controlled differentially private decentralized learning algorithm designed for a set of clients aiming to collaboratively train a common learning model. The network is characterized by a row-stochastic adjacency matrix, which reflects different channel gains between the clients. In our privacy-preserving approach, both the transmit power for model updates and the level of injected Gaussian noise are jointly controlled to satisfy a given privacy and energy budget. We show that our proposed algorithm achieves a convergence rate of O(log T), where T is the horizon bound in the regret function. Furthermore, our numerical results confirm that our proposed algorithm outperforms existing works.

Problem

• Develops differentially private decentralized learning for heterogeneous multicast networks
• Controls transmit power and Gaussian noise to meet privacy-energy constraints
• Achieves O(log T) convergence rate while outperforming existing methods

Innovation

• Power-controlled differentially private decentralized learning algorithm
• Joint control of transmit power and Gaussian noise injection
• Achieves O(log T) convergence rate with privacy guarantees