arxivstat.MLcs.LG2026-07-11
Integrating Background Knowledge for Scalable Causal Discovery
Mátyás Schubert, Theofanis Aslanidis, Tom Claassen, Sara Magliacane
Expert background knowledge is often available in practical applications of causal discovery. Such constraints on the true causal graph can help causal discovery in terms of identifiability of causal effects and accuracy of the learned structure, but also in reducing the space of…