Stabilizing Estimates of Shapley Values with Control Variates

arXiv:2310.07672v3 Announce Type: replace-cross
Abstract: Shapley values are among the most popular tools for explaining predictions of blackbox machine learning models. However, their high computational cost motivates the use of sampling approximations, inducing a considerable degree of uncertainty. To stabilize these model explanations, we propose ControlSHAP, an approach based on the Monte Carlo technique of control variates. Our methodology is applicable to any machine learning model and requires virtually no extra computation or modeling effort. On several high-dimensional datasets, we …

abstract arxiv blackbox computational control cost cs.lg however machine machine learning machine learning models popular predictions sampling stat.ml tools type uncertainty values