# QT-Routenet: Improved GNN generalization to larger 5G networks by fine-tuning predictions from queueing theory

arXiv: https://arxiv.org/abs/2207.06336 · HTML: https://mlsanity.com/paper/2207.06336/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 14 references that are in the map.
- First posted on arXiv 13 Jul 2022.

## Builds on (most cited first, 11 of 11)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [Fast Graph Representation Learning with PyTorch Geometric](https://mlsanity.com/paper/1903.02428/) (2019)
- [Structured Sequence Modeling with Graph Convolutional Recurrent Networks](https://mlsanity.com/paper/1612.07659/) (2016)
- [PyTorch Geometric Temporal: Spatiotemporal Signal Processing with Neural Machine Learning Models](https://mlsanity.com/paper/2104.07788/) (2021)
- [RouteNet: Leveraging Graph Neural Networks for network modeling and optimization in SDN](https://mlsanity.com/paper/1910.01508/) (2019)
- [Unveiling the potential of Graph Neural Networks for network modeling and optimization in SDN](https://mlsanity.com/paper/1901.08113/) (2019)
- [The Graph Neural Networking Challenge: A Worldwide Competition for Education in AI/ML for Networks](https://mlsanity.com/paper/2107.12433/) (2021)
- [Scaling Graph-based Deep Learning models to larger networks](https://mlsanity.com/paper/2110.01261/) (2021)
- [Optimizing Diffusion Rate and Label Reliability in a Graph-Based Semi-supervised Classifier](https://mlsanity.com/paper/2201.03456/) (2022)

## Built on by (most cited first, 2 of 2)

- [m4: A Learned Flow-level Network Simulator](https://mlsanity.com/paper/2503.01770/) (2025)
- [From Simulation to Deep Learning: Survey on Network Performance Modeling Approaches](https://mlsanity.com/paper/2603.28394/) (2026)

## Abstract (arXiv)

In order to promote the use of machine learning in 5G, the International Telecommunication Union (ITU) proposed in 2021 the second edition of the ITU AI/ML in 5G challenge, with over 1600 participants from 82 countries. This work details the second place solution overall, which is also the winning solution of the Graph Neural Networking Challenge 2021. We tackle the problem of generalization when applying a model to a 5G network that may have longer paths and larger link capacities than the ones observed in training. To achieve this, we propose to first extract robust features related to Queueing Theory (QT), and then fine-tune the analytical baseline prediction using a modification of the Routenet Graph Neural Network (GNN) model. The proposed solution generalizes much better than simply using Routenet, and manages to reduce the analytical baseline's 10.42 mean absolute percent error to 1.45 (1.27 with an ensemble). This suggests that making small changes to an approximate model that is known to be robust can be an effective way to improve accuracy without compromising generalization.