Double-Diffusion: Diffusion Conditioned Diffusion Probabilistic Model For Air Quality Prediction
This work addresses air quality prediction, a domain-specific problem with spatio-temporal complexity, by introducing a novel conditional generative approach using physics, which is incremental as it builds on existing diffusion models.
The paper tackles air quality prediction by proposing Double-Diffusion, a diffusion probabilistic model that uses known physics to guide forecasting with stochasticity, achieving a 3-12% increase in CRPS and reducing inference time by 30-50% compared to other probabilistic models.
Air quality prediction is a challenging forecasting task due to its spatio-temporal complexity and the inherent dynamics as well as uncertainty. Most of the current models handle these two challenges by applying Graph Neural Networks or known physics principles, and quantifying stochasticity through probabilistic networks like Diffusion models. Nevertheless, finding the right balancing point between the certainties and uncertainties remains an open question. Therefore, we propose Double-Diffusion, a novel diffusion probabilistic model that harnesses the power of known physics to guide air quality forecasting with stochasticity. To the best of our knowledge, while precedents have been made of using conditional diffusion models to predict air pollution, this is the first attempt to use physics as a conditional generative approach for air quality prediction. Along with a sampling strategy adopted from image restoration and a new denoiser architecture, Double-Diffusion ranks first in most evaluation scenarios across two real-life datasets compared with other probabilistic models, it also cuts inference time by 50% to 30% while enjoying an increase between 3-12% in Continuous Ranked Probabilistic Score (CRPS).