# Distributed Link Sparsification for Scalable Scheduling Using Graph Neural Networks (Journal Version)

arXiv: https://arxiv.org/abs/2509.05447 · HTML: https://mlsanity.com/paper/2509.05447/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 71 references that are in the map.
- First posted on arXiv 5 Sep 2025.

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

- [A Comprehensive Survey on Graph Neural Networks](https://mlsanity.com/paper/1901.00596/) (2019)
- [A Primer on Zeroth-Order Optimization in Signal Processing and Machine Learning](https://mlsanity.com/paper/2006.06224/) (2020)
- [Distributed Scheduling using Graph Neural Networks](https://mlsanity.com/paper/2011.09430/) (2020)
- [Link Scheduling using Graph Neural Networks](https://mlsanity.com/paper/2109.05536/) (2021)
- [Learning Decentralized Wireless Resource Allocations with Graph Neural Networks](https://mlsanity.com/paper/2107.01489/) (2021)
- [Massive Access for 5G and Beyond](https://mlsanity.com/paper/2002.03491/) (2020)
- [Towards Massive Machine Type Communications in Ultra-Dense Cellular IoT Networks: Current Issues and Machine Learning-Assisted Solutions](https://mlsanity.com/paper/1808.02924/) (2018)
- [The Internet of Battle Things](https://mlsanity.com/paper/1712.08980/) (2017)
- [Distributed Link Sparsification for Scalable Scheduling Using Graph Neural Networks](https://mlsanity.com/paper/2203.14339/) (2022)
- [Generalization of Graph Neural Networks is Robust to Model Mismatch](https://mlsanity.com/paper/2408.13878/) (2024)
- [Traffic-aware Two-stage Queueing Communication Networks: Queue Analysis and Energy Saving](https://mlsanity.com/paper/1909.05891/) (2019)

## Abstract (arXiv)

In wireless networks characterized by dense connectivity, the significant signaling overhead generated by distributed link scheduling algorithms can exacerbate issues like congestion, energy consumption, and radio footprint expansion. To mitigate these challenges, we propose a distributed link sparsification scheme employing graph neural networks (GNNs) to reduce scheduling overhead for delay-tolerant traffic while maintaining network capacity. A GNN module is trained to adjust contention thresholds for individual links based on traffic statistics and network topology, enabling links to withdraw from scheduling contention when they are unlikely to succeed. Our approach is facilitated by a novel offline constrained {unsupervised} learning algorithm capable of balancing two competing objectives: minimizing scheduling overhead while ensuring that total utility meets the required level. In simulated wireless multi-hop networks with up to 500 links, our link sparsification technique effectively alleviates network congestion and reduces radio footprints across four distinct distributed link scheduling protocols.