AirDDE: Multifactor Neural Delay Differential Equations for Air Quality Forecasting

2026-03-18

AI Summary

This work addresses the challenge of accurately modeling the inherent delays and complex spatiotemporal dynamics in air pollutant propagation, which existing prediction methods often fail to capture. To this end, we propose AirDDE, a novel framework that introduces neural delay differential equations (neural DDEs) into air quality forecasting for the first time. AirDDE integrates a physics-informed continuous-time evolution model grounded in advection–diffusion equations to explicitly represent delayed advection, diffusion, and source–sink effects of pollutants. Furthermore, it incorporates a memory-augmented attention mechanism that adaptively captures historical dependencies under multi-factor modulation. Evaluated on three real-world datasets, AirDDE achieves state-of-the-art performance, reducing the average mean absolute error (MAE) by 8.79% compared to the best baseline, while maintaining both high predictive accuracy and strong physical interpretability.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Diffusion Models for Vision
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Accurate air quality forecasting is essential for public health and environmental sustainability, but remains challenging due to the complex pollutant dynamics. Existing deep learning methods often model pollutant dynamics as an instantaneous process, overlooking the intrinsic delays in pollutant propagation. Thus, we propose AirDDE, the first neural delay differential equation framework in this task that integrates delay modeling into a continuous-time pollutant evolution under physical guidance. Specifically, two novel components are introduced: (1) a memory-augmented attention module that retrieves globally and locally historical features, which can adaptively capture delay effects modulated by multifactor data; and (2) a physics-guided delay evolving function, grounded in the diffusion-advection equation, that models diffusion, delayed advection, and source/sink terms, which can capture delay-aware pollutant accumulation patterns with physical plausibility. Extensive experiments on three real-world datasets demonstrate that AirDDE achieves the state-of-the-art forecasting performance with an average MAE reduction of 8.79\% over the best baselines. The code is available at https://github.com/w2obin/airdde-aaai.

Problem

• air quality forecasting
• pollutant dynamics
• time delay
• delay differential equations
• environmental modeling

Innovation

• Neural Delay Differential Equations
• Air Quality Forecasting
• Physics-Guided Modeling
• Memory-Augmented Attention
• Delay-Aware Dynamics