Abstract. While invariance of causal mechanisms has inspired recent work in both robust machine learning and causal inference, causal mechanisms often vary over domains due to, for example, population-specific differences, the context of data collection, or intervention. To discover invariant and changing mechanisms from data, we propose extending the algorithmic model for causation to mechanism changes and instantiating it via Minimum Description Length. In essence, for a continuous variable \(Y\) in multiple contexts \(C\), we identify variables \(X\) as causal if the regression functions \(g : X \rightarrow Y\) have succinct descriptions in all contexts. In empirical evaluations we show that our method, Vario, reveals mechanism changes, discovers causal variables by invariance, and finds causal networks, such as on real-world data that gives insight into the signaling pathways in human immune cells.

Update. Vario considers parametric models. With Linc, we can non-parametrically identify fully-oriented causal networks and mechanism changes.