Symposium Aiming to Improve the Climate

In the fall of 2024, Linköping University once again hosted ELLIIT’s five-week-long focus period. This ambitious guest researcher program aimed for greater breadth in interdisciplinarity this year, with the theme of machine learning for climate science.

The concept behind the focus periods started in the early 2010s at the Department of Automatic Control at Lund University. Lund, which is one of ELLIIT’s academic partners along with Linköping University, Halmstad University, and Blekinge Institute of Technology, later brought the concept into ELLIIT, where it was adopted as ELLIIT Focus Periods.

Machine Learning for Climate Science

The subject of this fall’s focus period is machine learning for climate science, taking place in Linköping between September 23rd and October 25th. The goal is to gather researchers from diverse backgrounds to lay the foundation for new research on how machine learning can be used to tackle critical problems in climate science – issues that currently lack satisfactory solutions.

Several international researchers interested in these research challenges have been invited to participate and engage with leading researchers, including ELLIIT researchers. During the third week of the focus period, the customary three-day symposium was held, featuring several invited top researchers in the field. The symposium is the highlight of the five-week focus period, with the goal of gathering researchers who share an interest in machine learning for climate science in general, and machine-learning-enhanced modeling and analysis of climate and weather systems in particular.

Relevant Theme

The organizers of this fall’s focus period at Linköping University are Fredrik Lindsten, Senior Associate Professor in Machine Learning, and Michael Felsberg, Professor in Computer Vision.

– “Many have come forward and said they appreciated the symposium, that there was a clear focus and that they were able to have beneficial scientific discussions with almost all participants. I wasn’t entirely sure that would be the case since we had participants clearly from ‘one side,’ either machine learning or climate, and we also covered quite a broad spectrum of issues within climate science. But it was clear that many were able to connect through a shared interest in using machine learning within this application area,” says Fredrik Lindsten.

Michael Felsberg concluded by emphasizing the relevance of the chosen focus period theme.

– “I think this focus period is working exceptionally well; the researchers’ backgrounds overlap more than I expected, and as a result, the discussions are intense and effective. Events in the world, the Nobel Prize in Physics and the hurricane in Florida, highlight the relevance and confirm the methodological choices we made in selecting the theme ‘AI for Climate.’”