Air Quality Prediction with Physics-Guided Dual Neural ODEs in Open Systems

2024-10-25

AI Summary

Air pollution forecasting faces dual challenges in open-system settings: the complexity of underlying physical mechanisms and the disregard for dynamical constraints by purely data-driven models. To address these, we propose a dual-branch Neural Ordinary Differential Equation (Neural ODE) framework. One branch explicitly incorporates governing equations of open-system physics to ensure interpretability and physical consistency; the other branch learns unmodeled processes in a data-driven manner. A spatiotemporal alignment mechanism dynamically fuses the two branches. This work is the first to achieve synergistic complementarity between explicit physical constraints and implicit neural learning under the open-system assumption, eliminating the mismatch between explicit equations and implicit representations prevalent in conventional physics-guided approaches. Evaluated on multi-scale air quality forecasting tasks, our method achieves state-of-the-art performance, with significant improvements in long-term prediction stability and physical plausibility.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Search and Optimization: Learning to Search
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Air pollution significantly threatens human health and ecosystems, necessitating effective air quality prediction to inform public policy. Traditional approaches are generally categorized into physics-based and data-driven models. Physics-based models usually struggle with high computational demands and closed-system assumptions, while data-driven models may overlook essential physical dynamics, confusing the capturing of spatiotemporal correlations. Although some physics-guided approaches combine the strengths of both models, they often face a mismatch between explicit physical equations and implicit learned representations. To address these challenges, we propose Air-DualODE, a novel physics-guided approach that integrates dual branches of Neural ODEs for air quality prediction. The first branch applies open-system physical equations to capture spatiotemporal dependencies for learning physics dynamics, while the second branch identifies the dependencies not addressed by the first in a fully data-driven way. These dual representations are temporally aligned and fused to enhance prediction accuracy. Our experimental results demonstrate that Air-DualODE achieves state-of-the-art performance in predicting pollutant concentrations across various spatial scales, thereby offering a promising solution for real-world air quality challenges.

Problem

• Predict air quality accurately using physics and data-driven models
• Address mismatch between physical equations and learned representations
• Improve spatiotemporal correlation capture in open systems

Innovation

• Physics-guided dual Neural ODEs integration
• Open-system physical equations for spatiotemporal dependencies
• Data-driven branch complements physics dynamics