A Variational U-Net for Weather Forecasting
This incremental work addresses weather prediction for climate and competition applications.
The authors tackled weather forecasting by developing a Variational U-Net that combines probabilistic modeling with fine detail recovery, achieving third place in the Weather4cast competition.
Not only can discovering patterns and insights from atmospheric data enable more accurate weather predictions, but it may also provide valuable information to help tackle climate change. Weather4cast is an open competition that aims to evaluate machine learning algorithms' capability to predict future atmospheric states. Here, we describe our third-place solution to Weather4cast. We present a novel Variational U-Net that combines a Variational Autoencoder's ability to consider the probabilistic nature of data with a U-Net's ability to recover fine-grained details. This solution is an evolution from our fourth-place solution to Traffic4cast 2020 with many commonalities, suggesting its applicability to vastly different domains, such as weather and traffic.