Toward Causal Learning in Real World Applications: Advances in Nonlinear and Time Series Models (CausalANTS)

In causality we want to understand how a system reacts under interventions (e.g., in gene knock-out experiments). These questions go beyond statistical dependences and cannot be answered by standard methods from statistics. Causal models are stable with respect to changes in the environment, a property that will help us to learn causal structures from data. Together with two PhD students and one postdoc we will apply our research in earth system science, biogeography, and computational biology.