HypNO: A Graph-Based Neural Operator with Physics-Informed Message Passing for Hyperbolic Conservation Laws
Machine Learning
2026-07-10 v1 Artificial Intelligence
Abstract
We introduce HypNO, a graph-based neural operator for scalar hyperbolic conservation laws. HypNO operates directly on a space-time graph of finite-volume cells and uses adjacency-factored, physics-informed message passing to respect upwinding and entropy admissibility near shocks. We benchmark the architecture on the Lighthill-Whitham-Richards (LWR) and Aw-Rascle-Zhang (ARZ) traffic-flow models, a stress test for operator-learning methods because of their simultaneous global transport and shock formation. HypNO predicts solution snapshots accurately across a range of initial conditions while capturing the shocks and discontinuities of the solution.
Cite
@article{arxiv.2607.20541,
title = {HypNO: A Graph-Based Neural Operator with Physics-Informed Message Passing for Hyperbolic Conservation Laws},
author = {Dimitrije Ždrale and Cassie An Jeng and Katie Wang and Sonia Vanier and Alexandre Bayen and Hossein Nick Zinat Matin},
journal= {arXiv preprint arXiv:2607.20541},
year = {2026}
}