AI RESEARCH
cc-Shapley: Measuring Multivariate Feature Importance Needs Causal Context
arXiv CS.LG
•
ArXi:2602.20396v3 Announce Type: replace Explainable artificial intelligence promises to yield insights into relevant features, thereby enabling humans to examine and scrutinize machine learning models or even facilitating scientific discovery. Considering the widespread technique of Shapley values, we find that purely data-driven operationalization of multivariate feature importance is unsuitable for such purposes.