On the Optimization of Model Aggregation for Federated Learning at the Network Edge

2025-11-04

IEEE Transactions on Network and Service Management

AI Summary

To address severe cloud-central link congestion, high training failure rates, and underutilized edge computing resources in edge-based federated learning, this paper proposes a novel hierarchical aggregation architecture featuring edge-intermediate aggregation. By integrating Multi-access Edge Computing (MEC) with Software-Defined Wide Area Networking (SD-WAN), we construct an aggregator coverage network. We formulate an Integer Linear Programming (ILP) model to jointly capture the coupling among aggregation routing decisions, resource constraints, and communication overhead. Furthermore, we design an efficient heuristic algorithm for dynamic aggregation path optimization. Experimental results demonstrate that, while preserving model accuracy, the proposed approach reduces per-round training failure rates by up to 15%, significantly alleviates cloud-link congestion, improves network resource utilization, and enhances system robustness. To the best of our knowledge, this is the first work to achieve joint scheduling optimization of federated aggregation in SD-WAN-enhanced MEC environments.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Multiagent Systems: Mechanism Design
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

The rapid increase in connected devices has signifi- cantly intensified the computational and communication demands on modern telecommunication networks. To address these chal- lenges, integrating advanced Machine Learning (ML) techniques like Federated Learning (FL) with emerging paradigms such as Multi-access Edge Computing (MEC) and Software-Defined Wide Area Networks (SD-WANs) is crucial. This paper intro- duces online resource management strategies specifically designed for FL model aggregation, utilizing intermediate aggregation at edge nodes. Our analysis highlights the benefits of incorporating edge aggregators to reduce network link congestion and maximize the potential of edge computing nodes. However, the risk of network congestion persists. To mitigate this, we propose a novel aggregation approach that deploys an aggregator overlay network. We present an Integer Linear Programming (ILP) model and a heuristic algorithm to optimize the routing within this overlay network. Our solution demonstrates improved adapt- ability to network resource utilization, significantly reducing FL training round failure rates by up to 15% while also alleviating cloud link congestion.

Problem

• Optimizing model aggregation for Federated Learning at network edge
• Reducing network link congestion through edge aggregator deployment
• Minimizing FL training failures via overlay network routing optimization

Innovation

• Intermediate aggregation at edge nodes
• Aggregator overlay network for congestion mitigation
• ILP model and heuristic for routing optimization