We humans are used to recognizing patterns in the world around us and we explain these patterns in complex relationships of cause and effect. Our everyday lives are increasingly shaped by artificial intelligence. Large amounts of data are generated every day in our digitalized world. But it is still a major challenge for artificial intelligence to identify causal relationships in this data. This is especially true when we are not able to experiment with the system. Prof. Dr. Jakob Runge and his research group "Causal Inference" at the DLR Institute of Data Science are developing new theoretical foundations and the associated tools to help scientists at DLR and beyond to extract causal relationships from data. An important basic problem of causal inference is to quantify the causal effects as precisely as possible from purely qualitative knowledge about causal dependencies in the form of causal networks.
Jakob Runge has now been awarded the DLR Science Prize. In his award-winning work, Jakob Runge developed a new theoretical approach and a method derived from it that allows causal relationships to be calculated very precisely, even taking unobserved processes into account. The work was published at the most important conference for artificial intelligence and machine learning "NeurIPS". Causal knowledge allows us to better predict the world around us and answer one of the simplest questions of all: Why?
