Built independently by an author, for readers. Read the story and support ChapterPal
ChapterPal Logo
Log inSign up

keyword

Graph Neural Controlled Differential Equations

Graph Neural Controlled Differential Equations are continuous-time deep learning models designed to process and forecast dynamic, sequential data structured as graphs. By extending the mathematical framework of neural controlled differential equations to network topologies, these architectures model system states as continuous trajectories driven by continuous input paths constructed from discrete or irregularly sampled observations. The framework typically integrates graph neural network operations with controlled differential equation solvers to simultaneously capture temporal dynamics and spatial interdependencies among interconnected nodes. This formulation allows the model to process irregular time intervals and missing data naturally without structural alterations, making it widely applicable to spatio-temporal forecasting tasks such as traffic prediction and dynamical physical system modeling.

1 item

Graph Neural Controlled Differential Equations for Traffic Forecasting

Graph Neural Controlled Differential Equations for Traffic Forecasting

Jeongwhan Choi, Hwangyong Choi, Jeehyun Hwang, Noseong Park

OrganizationsYonsei University

Why you should read this

Develops a unified spatio-temporal neural controlled differential equation framework that continuously models both spatial graph dynamics and temporal traffic patterns, significantly outperforming existing baselines across benchmark forecasting datasets.

Traffic forecasting is one of the most popular spatio-temporal tasks in the field of machine learning. A prevalent approach in the field is to combine graph convolutional networks and recurrent neural networks for the spatio-temporal processing. There has been fierce competition and many novel methods have been proposed. In this paper, we present the method of spatio-temporal graph neural controlled differential equation (STG-NCDE). Neural controlled differential equations (NCDEs) are a breakthrough concept for processing sequential data. We extend the concept and design two NCDEs: one for the temporal processing and the other for the spatial processing. After that, we combine them into a single framework. We conduct experiments with 6 benchmark datasets and 20 baselines. STG-NCDE shows the best accuracy in all cases, outperforming all those 20 baselines by non-trivial margins.

Added

2026-09-26