Shapley Curves: A Smoothing Perspective

arXiv:2211.13289v5 Announce Type: replace-cross
Abstract: This paper fills the limited statistical understanding of Shapley values as a variable importance measure from a nonparametric (or smoothing) perspective. We introduce population-level \textit{Shapley curves} to measure the true variable importance, determined by the conditional expectation function and the distribution of covariates. Having defined the estimand, we derive minimax convergence rates and asymptotic normality under general conditions for the two leading estimation strategies. For finite sample inference, we propose a novel version of the …

abstract arxiv cs.lg distribution function importance minimax paper perspective population statistical stat.me stat.ml true type understanding values