MACHINE LEARNING FOR SPACE SCIENCE: PROJECTS AND COMMUNITY BUILDING - SANDOR KRUK

In the past year, our Data Science team has dedicated efforts to foster a community centered on using machine learning in space science. We have actively involved young scientists at ESA in various activities on ESA Datalabs and are building knowledge and expertise through a range of projects, from natural language processing to computer vision. In this presentation, I will highlight some of our current projects, such as fine-tuning for the first time a large language model for astronomical research. I will discuss the role of infrastructures such as ESA Datalabs and discuss the scientific networks that have emerged throughout these ML initiatives, such as UniverseTBD.