RKHS-SHAP: Shapley Values for Kernel MethodsDownload PDF

Published: 31 Oct 2022, 18:00, Last Modified: 10 Jan 2023, 19:00NeurIPS 2022 AcceptReaders: Everyone
Keywords: Kernel Methods, Shapley Values, Explainable AI
TL;DR: Proposing a kernel-method specific SHAP algorithm for RKHS model explainability and analyse its theoretical and practical properties.
Abstract: Feature attribution for kernel methods is often heuristic and not individualised for each prediction. To address this, we turn to the concept of Shapley values (SV), a coalition game theoretical framework that has previously been applied to different machine learning model interpretation tasks, such as linear models, tree ensembles and deep networks. By analysing SVs from a functional perspective, we propose RKHS-SHAP, an attribution method for kernel machines that can efficiently compute both Interventional and Observational Shapley values using kernel mean embeddings of distributions. We show theoretically that our method is robust with respect to local perturbations - a key yet often overlooked desideratum for consistent model interpretation. Further, we propose Shapley regulariser, applicable to a general empirical risk minimisation framework, allowing learning while controlling the level of specific feature's contributions to the model. We demonstrate that the Shapley regulariser enables learning which is robust to covariate shift of a given feature and fair learning which controls the SVs of sensitive features.
Supplementary Material: pdf
24 Replies

Loading