Reliable Vertical Federated Learning in 5G Core Network Architecture

--This work proposes a new algorithm to mitigate model generalization loss in V ertical 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. In the realm of 5G networks, there is a growing emphasis on integrating machine learning frameworks into daily operations for both edge devices and operators, driven by the proven advantages these frameworks can offer over traditional model-based methods. Among these, Federated Learning (FL) [1] has gained prominence due to its distributed and privacy-preserving framework during both training and inference.