UQGNN: Uncertainty Quantification of Graph Neural Networks for Multivariate Spatiotemporal Prediction

TL;DR

UQGNN is a novel graph neural network model that jointly predicts multivariate spatiotemporal data and quantifies uncertainty, improving reliability and accuracy in urban applications.

Contribution

It introduces a multivariate probabilistic GNN with interaction-aware embedding and uncertainty estimation, addressing the gap in modeling correlated urban phenomena.

Findings

Outperforms state-of-the-art models in accuracy and uncertainty quantification.

Achieves 5% improvement on Shenzhen dataset in both metrics.

Demonstrates effectiveness across multiple real-world datasets.

Abstract

Spatiotemporal prediction plays a critical role in numerous real-world applications such as urban planning, transportation optimization, disaster response, and pandemic control. In recent years, researchers have made significant progress by developing advanced deep learning models for spatiotemporal prediction. However, most existing models are deterministic, i.e., predicting only the expected mean values without quantifying uncertainty, leading to potentially unreliable and inaccurate outcomes. While recent studies have introduced probabilistic models to quantify uncertainty, they typically focus on a single phenomenon (e.g., taxi, bike, crime, or traffic crashes), thereby neglecting the inherent correlations among heterogeneous urban phenomena. To address the research gap, we propose a novel Graph Neural Network with Uncertainty Quantification, termed UQGNN for multivariate spatiotemporal prediction. UQGNN introduces two key innovations: (i) an Interaction-aware Spatiotemporal Embedding Module that integrates a multivariate diffusion graph convolutional network and an interaction-aware temporal convolutional network to effectively capture complex spatial and temporal interaction patterns, and (ii) a multivariate probabilistic prediction module designed to estimate both expected mean values and associated uncertainties. Extensive experiments on four real-world multivariate spatiotemporal datasets from Shenzhen, New York City, and Chicago demonstrate that UQGNN consistently outperforms state-of-the-art baselines in both prediction accuracy and uncertainty quantification. For example, on the Shenzhen dataset, UQGNN achieves a 5% improvement in both prediction accuracy and uncertainty quantification.