XAI-Driven Client Selection for Federated Learning in Scalable 6G Network Slicing

TL;DR

The paper tackles scalable, trustworthy federated learning for 6G network slicing under non-IID data by introducing an XAI-guided client selection mechanism. It leverages Integrated Gradients attributions to prioritize a subset of Analytic Engines (AEs) in each FL round, enabling faster convergence and lower computation/communication overhead via FedAvg. Experiments across multiple slices demonstrate improved training efficiency and scalability, with favorable comparisons to no-policy and score-based baselines. The approach enhances transparency and robustness in zero-touch network automation, with promising directions toward cloud-native deployments and live-edge testbeds.

Abstract

In recent years, network slicing has embraced artificial intelligence (AI) models to manage the growing complexity of communication networks. In such a situation, AI-driven zero-touch network automation should present a high degree of flexibility and viability, especially when deployed in live production networks. However, centralized controllers suffer from high data communication overhead due to the vast amount of user data, and most network slices are reluctant to share private data. In federated learning systems, selecting trustworthy clients to participate in training is critical for ensuring system performance and reliability. The present paper proposes a new approach to client selection by leveraging an XAI method to guarantee scalable and fast operation of federated learning based analytic engines that implement slice-level resource provisioning at the RAN-Edge in a non-IID scenario. Attributions from XAI are used to guide the selection of devices participating in training. This approach enhances network trustworthiness for users and addresses the black-box nature of neural network models. The simulations conducted outperformed the standard approach in terms of both convergence time and computational cost, while also demonstrating high scalability.

Figures (7)

• Figure 1: Proposed architecture.
• Figure 2: Proposed policy for AE selection.
• Figure 3: FL training MSE loss vs. number of FL rounds for $m=5$ and $K=10$.
• Figure 4: Completion time vs. number of FL rounds for $m=5$ and $K=10$.
• Figure 5: FL training MSE loss vs. number of FL rounds with proposed policy for $m=25$ and $K=(40,50)$.