Recently appointed as Professor at Linköping University, Fredrik Lindsten is one of the university’s leading experts in AI technology and machine learning. We spoke to him about organizing focus periods, machine learning in weather forecasting, and his ambitions within ELLIIT.
Fredrik Lindsten, Professor of Machine Learning and Head of the Division of Statistics and Machine Learning at Linköping University, joined the ELLIIT Recruited Faculty in 2020. Since then, Lindsten has been active in multiple ELLIIT activities, not least as one of the organizers of the ELLIIT Focus Period on Machine Learning for Climate Science in late 2024.
As one the organizers of an ELLIIT Focus Period, what can you tell us about that experience?
It was a very nice experience that allowed me to connect with many researchers in this area that I most likely would not have met otherwise. In addition to being a fun and engaging event, I believe that it has contributed to pushing the research frontier forward. We are now starting to see cross-disciplinary publications coming out as a result of the collaborations that were initiated during the focus period.
Please, tell us about your own research and projects related to ELLIIT.
I don’t have any projects from the research project calls, but I have a lot of ongoing research activities related to ELLIIT. Lately, a focus area has been spatio-temporal and generative machine learning models, in particular with applications in machine learning weather forecasting. Among other things, we have developed probabilistic (generative) models for quantifying uncertainty in the forecasts, worked on improving the temporal resolution, designing regional machine-learning-based weather forecasting, and developing new data assimilation methods based on generative models. I also work on machine learning for materials science, where we also use generative models for simulating hypothetical materials with desirable properties.
Where did you find your interest in this research area?
I’ve always been interested in mathematics and in particular probability theory and statistics. Working with generative machine learning for scientific applications allows me to combine interesting mathematics with applied problems of high societal relevance. Working on machine learning has also allowed me to engage in many diverse interdisciplinary collaborations which I find very rewarding.
Is there anything else you’re working on that you’d like to highlight?
Perhaps work linking to last years focus period, and ongoing collaboration with SMHI. Tomas Landelius at SMHI is currently an Adjunct Senior Associate Professor at my division.
What are your ambitions with ELLIIT in the future?
To further strengthen Sweden’s position in AI and machine learning research in areas of high societal and industrial relevance. ELLIIT is providing excellent conditions for this, both in terms of research funding, but also through other initiatives such as the aforementioned focus period. Initiatives of this type can act as a catalyst for new collaborations and for identifying interesting and relevant problem formulations.
And finally, what do you enjoy doing in your free time outside of academia?
I have three kids, aged 5, 8, and 10, and enjoy spending time with family and tagging along on their activities. When I find some free time for myself I enjoy running (I’m also a wannabe orienteering runner), playing boardgames and listening to metal music.
Photos: Magnus Johansson (LiU).
Contact and more
Read more about Fredrik Lindsten’s research and projects, and find contact information here.
Previous Meet the Recruited Faculty
See more interviews from our series Meet the Recruited Faculty here.
Machine Learning for Climate Science
Read more about the ELLIIT Focus Period on Machine Learning for Climate Science.