AirCast通过多变量对齐与加权损失，提升极端污染事件预测精度。
AirCast: Improving Air Pollution Forecasting Through Multi-Variable Data Alignment
- 多任务学习融合气象与污染物数据，捕捉天气对污染影响。
- 引入频率加权MAE损失，改善罕见极端污染事件的预测效果。
- 基于领域知识筛选关键变量，适合环境监测与公共健康研究者。
空气污染仍是全球主要健康风险，受工业化和城市化加剧，显著增加发病率与死亡率。本文提出AirCast，一种新型多变量空气质量预测模型，整合气象与空气质量变量。该模型采用多任务头架构，同步预测大气状况与污染物浓度，增强对天气影响空气质量的理解。由于历史数据中极端污染事件稀少，污染水平呈现重尾分布，预测难度大。为此，我们提出适用于回归任务的频次加权平均绝对误差（fMAE）损失，借鉴类别平衡损失思想。结合领域知识，筛选影响污染的关键变量，并在时空维度上对齐现有气象与化学数据集。AirCast通过多任务学习、频率加权损失及领域驱动变量选择的综合方法，实现更精准的污染预测。代码与模型已开源（https://github.com/vishalned/AirCast.git）。
原文摘要 · Abstract (English)
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)
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。