From GNNs to Symbolic Surrogates via Kolmogorov-Arnold Networks for Delay Prediction
This addresses the need for efficient and transparent delay prediction models in communication network optimization, though it is incremental as it builds on existing GNN and KAN methods.
The paper tackled flow delay prediction in communication networks by developing FlowKANet, which replaces standard MLP layers with Kolmogorov-Arnold Networks to reduce trainable parameters while maintaining competitive performance, and then distilled it into symbolic surrogate models for lightweight deployment.
Accurate prediction of flow delay is essential for optimizing and managing modern communication networks. We investigate three levels of modeling for this task. First, we implement a heterogeneous GNN with attention-based message passing, establishing a strong neural baseline. Second, we propose FlowKANet in which Kolmogorov-Arnold Networks replace standard MLP layers, reducing trainable parameters while maintaining competitive predictive performance. FlowKANet integrates KAMP-Attn (Kolmogorov-Arnold Message Passing with Attention), embedding KAN operators directly into message-passing and attention computation. Finally, we distill the model into symbolic surrogate models using block-wise regression, producing closed-form equations that eliminate trainable weights while preserving graph-structured dependencies. The results show that KAN layers provide a favorable trade-off between efficiency and accuracy and that symbolic surrogates emphasize the potential for lightweight deployment and enhanced transparency.