InSlicing: Interpretable Learning-Assisted Network Slice Configuration in Open Radio Access Networks

TL;DR

InSlicing introduces an interpretable, learning-assisted algorithm for network slice configuration in 5G RANs, combining Kolmogorov-Arnold Networks and hybrid optimization to adapt dynamically and reduce operational costs.

Contribution

The paper presents a novel interpretable algorithm that integrates KANs with hybrid optimization for efficient network slice configuration in open RANs, addressing adaptability and interpretability.

Findings

Achieves high interpretability in slice configuration.

Reduces operation costs by over 25%.

Effectively adapts to dynamic network environments.

Abstract

Network slicing is a key technology enabling the flexibility and efficiency of 5G networks, offering customized services for diverse applications. However, existing methods face challenges in adapting to dynamic network environments and lack interpretability in performance models. In this paper, we propose a novel interpretable network slice configuration algorithm (\emph{InSlicing}) in open radio access networks, by integrating Kolmogorov-Arnold Networks (KANs) and hybrid optimization process. On the one hand, we use KANs to approximate and learn the unknown performance function of individual slices, which converts the blackbox optimization problem. On the other hand, we solve the converted problem with a genetic method for global search and incorporate a trust region for gradient-based local refinement. With the extensive evaluation, we show that our proposed algorithm achieves high interpretability while reducing 25+\% operation cost than existing solutions.