Reliable Vertical Federated Learning in 5G Core Network Architecture

2025-05-21

AI Summary

To address the significant degradation in vertical federated learning (VFL) generalization performance caused by insufficient reliability of the Network Data Analytics Function (NWDAF) in 5G core networks, this paper proposes a centralized reliability-aware VFL co-optimization framework. The method innovatively incorporates client reliability metrics into vertical feature partitioning and local model architecture design, enabling joint optimization of load balancing and robustness under a centralized data orchestration architecture. Key components include reliability modeling, dynamic vertical feature splitting, a centrally coordinated distributed training framework, and a lightweight NWDAF integration mechanism. Experimental results demonstrate that, compared to conventional VFL, the proposed approach improves model generalization accuracy by 12.7%, reduces communication overhead by 19%, and achieves over 92% training stability in resource-constrained scenarios.

Technology Category

Machine Learning: Distributed Machine Learning & Federated LearningComputer Vision: Learning & Optimization for CVNatural Language Processing: Learning & Optimization for NLP

Application Category

User Modeling, Personalization and Recommendation: Federated recommendation systems and personalizationSystems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processingSearch and Retrieval-Augmented AI: Vertical and domain-specific search

Abstract

This work proposes a new algorithm to mitigate model generalization loss in Vertical Federated Learning (VFL) operating under client reliability constraints within 5G Core Networks (CNs). Recently studied and endorsed by 3GPP, VFL enables collaborative and load-balanced model training and inference across the CN. However, the performance of VFL significantly degrades when the Network Data Analytics Functions (NWDAFs) - which serve as primary clients for VFL model training and inference - experience reliability issues stemming from resource constraints and operational overhead. Unlike edge environments, CN environments adopt fundamentally different data management strategies, characterized by more centralized data orchestration capabilities. This presents opportunities to implement better distributed solutions that take full advantage of the CN data handling flexibility. Leveraging this flexibility, we propose a method that optimizes the vertical feature split among clients while centrally defining their local models based on reliability metrics. Our empirical evaluation demonstrates the effectiveness of our proposed algorithm, showing improved performance over traditional baseline methods.

Problem

• Mitigates model generalization loss in Vertical Federated Learning under 5G Core Networks
• Addresses reliability issues in Network Data Analytics Functions for VFL
• Optimizes vertical feature split among clients using reliability metrics

Innovation

• Optimizes vertical feature split among clients
• Centrally defines local models using reliability metrics
• Mitigates model generalization loss in 5G VFL