Adversarial Node Placement in Decentralized Federated Learning: Maximum Spanning-Centrality Strategy and Performance Analysis

2025-01-01

IEEE Internet of Things Journal

AI Summary

This work identifies and systematically evaluates an overlooked security vulnerability in decentralized federated learning (FL): adversarial node placement. We propose MaxSpAN-FL, a hybrid attack strategy that jointly leverages maximum-span selection and network centrality analysis—specifically eigenvector centrality—and provide theoretical justification for its near-optimality in this context. To enhance practicality, we further design a topology-aware probabilistic deployment mechanism. Extensive experiments across diverse network topologies demonstrate that MaxSpAN-FL induces the most severe degradation in model convergence and final accuracy—substantially outperforming existing baseline attacks. Our approach offers both a novel analytical perspective on adversarial robustness in decentralized FL and a potent, general-purpose evaluation tool for security assessment.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

As federated learning (FL) becomes more widespread, there is growing interest in its decentralized variants. Decentralized FL leverages the benefits of fast and energy-efficient device-to-device communications to obviate the need for a central server. However, this opens the door to new security vulnerabilities as well. While FL security has been a popular research topic, the role of adversarial node placement in decentralized FL remains largely unexplored. This article addresses this gap by evaluating the impact of various coordinated adversarial node placement strategies on decentralized FL’s model training performance. We adapt two threads of placement strategies to this context: 1) maximum span-based algorithms and 2) network centrality-based approaches. Building on them, we propose a novel attack strategy, MaxSpAN-FL, which is a hybrid between these paradigms that adjusts node placement probabilistically based on network topology characteristics. Numerical experiments demonstrate that our attack consistently induces the largest degradation in decentralized FL models compared with baseline schemes across various network configurations and numbers of coordinating adversaries. We also provide theoretical support for why eigenvector centrality-based attacks are suboptimal in decentralized FL. Overall, our findings provide valuable insights into the vulnerabilities of decentralized FL systems, setting the stage for future research aimed at developing more secure and robust decentralized FL frameworks.

Problem

• Evaluating adversarial node placement impact on decentralized federated learning performance
• Proposing hybrid attack strategy combining maximum span and centrality approaches
• Analyzing vulnerabilities to develop more secure decentralized FL frameworks

Innovation

• Probabilistic node placement based on network topology
• Hybrid maximum span and centrality attack strategy
• Theoretical analysis of eigenvector centrality limitations