Shapley values for feature selection: The good, the bad, and the axioms
This work highlights a critical flaw in applying Shapley values to feature selection, which is important for researchers and practitioners in Explainable AI.
The paper critiques the use of Shapley values for feature selection, showing through counterexamples that their axioms can conflict with selection goals, and investigates this in simulations with SHAP and SAGE.
The Shapley value has become popular in the Explainable AI (XAI) literature, thanks, to a large extent, to a solid theoretical foundation, including four "favourable and fair" axioms for attribution in transferable utility games. The Shapley value is provably the only solution concept satisfying these axioms. In this paper, we introduce the Shapley value and draw attention to its recent uses as a feature selection tool. We call into question this use of the Shapley value, using simple, abstract "toy" counterexamples to illustrate that the axioms may work against the goals of feature selection. From this, we develop a number of insights that are then investigated in concrete simulation settings, with a variety of Shapley value formulations, including SHapley Additive exPlanations (SHAP) and Shapley Additive Global importancE (SAGE).