Causal models represent how changes in one part of a system affect other system components. In contrast to correlations in statistical (machine learning) models, they can thus explain why effects occur and how external interventions would impact the system. Causal models are grounded in expert knowledge, assumptions and data-based inferences and can ideally be tested and verified through experimentation to answer question of “what if”. They can be used to plot an optimal course of action and to reason counterfactually about what would have happened in case of alternative actions in the past to explain current and alternative outcomes, analyse errors or support predictions of the effect of future actions.

Contact: Jonas Wahl, André Meyer-Vitali