AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment

2025-02-25

AI Summary

Addressing the challenge of forecasting extreme air pollution events—exacerbated by their historical sparsity—the paper proposes a multi-task spatiotemporal fusion framework. Methodologically, it jointly models meteorological variables and multi-pollutant concentrations to enable cross-physical-process collaborative learning; introduces a frequency-weighted mean absolute error (fMAE) loss function to explicitly mitigate long-tail distribution bias in critical pollutants (e.g., PM₂.₅ and O₃); and incorporates domain-knowledge-driven key-variable selection and multi-source spatiotemporal data alignment. Empirically, the framework significantly improves both extreme-event detection capability and overall forecasting accuracy across short- to medium-term horizons. The source code and pre-trained models are publicly released.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Computer Vision: Multi-modal Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Air pollution remains a leading global health risk, exacerbated by rapid industrialization and urbanization, contributing significantly to morbidity and mortality rates. In this paper, we introduce AirCast, a novel multi-variable air pollution forecasting model, by combining weather and air quality variables. AirCast employs a multi-task head architecture that simultaneously forecasts atmospheric conditions and pollutant concentrations, improving its understanding of how weather patterns affect air quality. Predicting extreme pollution events is challenging due to their rare occurrence in historic data, resulting in a heavy-tailed distribution of pollution levels. To address this, we propose a novel Frequency-weighted Mean Absolute Error (fMAE) loss, adapted from the class-balanced loss for regression tasks. Informed from domain knowledge, we investigate the selection of key variables known to influence pollution levels. Additionally, we align existing weather and chemical datasets across spatial and temporal dimensions. AirCast's integrated approach, combining multi-task learning, frequency weighted loss and domain informed variable selection, enables more accurate pollution forecasts. Our source code and models are made public here (https://github.com/vishalned/AirCast.git)

Problem

• Improving air pollution forecasting accuracy
• Addressing rare extreme pollution events
• Aligning weather and chemical datasets

Innovation

• Multi-task head architecture
• Frequency-weighted Mean Absolute Error
• Domain-informed variable selection