AI RESEARCH
The Causal Description Gap: Information-Theoretic Separations Across Pearl's Hierarchy
arXiv CS.LG
•
ArXi:2605.02177v1 Announce Type: cross Pearl's causal hierarchy shows that observational, interventional, and counterfactual queries are qualitatively distinct. We ask a quantitative version of this question: how many additional bits are needed to specify higher-rung causal answers once lower-rung answers are known? We formalize this via query-class description length, the Kolmogoro complexity of the answer oracle induced by an SCM for a class of queries.