Machine Learning in Extreme Weather Forecasting with William Drew Collins

Learn how to apply machine learning in extreme weather forecasting to study rare and impactful weather events in a warming world. The course introduces the FourCastNet ML algorithm, which utilizes Fourier Neural Operators and Transformers to model chaotic dynamical systems like turbulent flows and atmospheric dynamics. By leveraging machine learning, hindcasts achieve high accuracy and fidelity at a significantly lower computational cost compared to traditional methods, enabling the generation of large ensembles. The teaching method includes a kickoff presentation followed by a Q&A session. This course is intended for individuals interested in the intersection of machine learning and geosciences, particularly in the field of extreme weather forecasting.