Learning from limited data in Earth system sciences: optimal selection of datasets and algorithms

x. November 2025
Prof. in Dr.in Nikki Vercauteren & Dr.in Claudia Acquistapace

When the underlying learning task is highly underdetermined, due to a large problem dimension and relatively small data statistics size. The small data challenges are particularly common in Earth System Sciences. The focus should be on mathematically transparent formulations and rigorous characterizations of the methods, so that theoretically founded interpretation of the methods and their results is possible. Discussions should be around highly efficient approaches to the small data problem and overfitting challenges in AI, and on the identification of key challenges motivated by AI application in Earth system sciences.

The main goal of this workshop is to bring together experts of DL mathematics and algorithmic with experts of Earth system sciences that use and possibly develop such methods.