Lessons and Outlook for ML Parameterization of Sub Grid Atmospheric Physics From the Vantage of Emulating Cloud Superparameterization - Mike Pritchard

This course focuses on the use of Machine Learning (ML) for parameterizing sub grid atmospheric physics, specifically in emulating cloud superparameterization. The learning outcomes include understanding the potential of big data and ML algorithms in studying climate systems, enabling climate scientists to ask causal questions and develop new theories. The course covers topics such as global modeling, multiskill modeling, GPU computing, feature engineering, neural network tuning, and more. The teaching method involves a conference setting aimed at exchanging tools and ideas, identifying key problems, and fostering collaborative efforts. The intended audience includes experts in earth system and computational sciences involved in addressing climate change challenges.