Topology-aware Neural Flux Prediction Guided by Physics

TL;DR

This paper introduces a topology-aware neural flux prediction framework that enhances GNNs' ability to model flow dynamics on directed graphs by incorporating explicit directional differences and physical constraints.

Contribution

It proposes a novel GNN framework combining difference matrices and physical constraints to better capture high-frequency topological features in directed graphs.

Findings

Improved flux prediction accuracy on water and traffic networks.

Enhanced sensitivity to directional and topological differences.

Effective modeling of flow dynamics in directed graphs.

Abstract

Graph Neural Networks (GNNs) often struggle in preserving high-frequency components of nodal signals when dealing with directed graphs. Such components are crucial for modeling flow dynamics, without which a traditional GNN tends to treat a graph with forward and reverse topologies equal.To make GNNs sensitive to those high-frequency components thereby being capable to capture detailed topological differences, this paper proposes a novel framework that combines 1) explicit difference matrices that model directional gradients and 2) implicit physical constraints that enforce messages passing within GNNs to be consistent with natural laws. Evaluations on two real-world directed graph data, namely, water flux network and urban traffic flow network, demonstrate the effectiveness of our proposal.