In his research, Prof. Runge develops theory, methods, and software for causal inference – with a strong application focus. Causal inference complements classical methods of statistics, machine learning, and artificial intelligence by going beyond mere correlations and enabling the learning of cause-and-effect relationships from data. Causal inference thus contributes to process understanding in many data-driven fields, from Earth system science and neuroscience to economics and urban studies.
Causal inference also lays the foundation for robust, interpretable, and scientifically sound AI systems that not only enable predictions under previously unseen conditions, but can also explain complex relationships and justify decisions.