Current Visitors:


Airphynet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction

KH. Hettige, J. Ji, S. Xiang, C. Long, G. Cong and J. Wang

in Proceedings of the Twelfth International Conference on Learning Representations (ICLR'24)


Air quality prediction and modelling plays a pivotal role in public health and en- vironment management, for individuals and authorities to make informed deci- sions. Although traditional data-driven models have shown promise in this do- main, their long-term prediction accuracy can be limited, especially in scenarios with sparse or incomplete data and they often rely on black-box deep learning structures that lack solid physical foundation leading to reduced transparency and interpretability in predictions. To address these limitations, this paper presents a novel approach named Physics guided Neural Network for Air Quality Prediction (AirPhyNet). Specifically, we leverage two well-established physics principles of air particle movement (diffusion and advection) by representing them as differ- ential equation networks. Then, we utilize a graph structure to integrate physics knowledge into a neural network architecture and exploit latent representations to capture spatio-temporal relationships within the air quality data. Experiments on two real-world benchmark datasets demonstrate that AirPhyNet outperforms state-of-the-art models for different testing scenarios including different lead time (24h, 48h, 72h), sparse data and sudden change prediction, achieving reduction in prediction errors up to 10%. Moreover, a case study further validates that our model captures underlying physical processes of particle movement and generates accurate predictions with real physical meaning.

Airphynet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction
Airphynet Harnessing Physics-Guided Neur
Adobe Acrobat Document 7.2 MB

@inproceedings{hettige2024airphynet,

title={AirPhyNet: Harnessing Physics-Guided Neural Networks for Air Quality Prediction}, 

author={Hettige, Kethmi Hirushini and Ji, Jiahao and Xiang, Shili and Long, Cheng and Cong, Gao and Wang, Jingyuan},

booktitle={Proceedings of the Twelfth International Conference on Learning Representations},

year={2024}