Ensembles of Random SHAPs

TL;DR

This paper introduces ensemble-based modifications to the SHAP method, aiming to reduce computational costs for local explanations of models with many features by approximating SHAP with smaller ensembles.

Contribution

The paper proposes three novel ensemble modifications of SHAP—ER-SHAP, ERW-SHAP, and ER-SHAP-RF—that improve efficiency and diversity in local explanations.

Findings

ER-SHAP reduces computation by averaging over small SHAPs.

ERW-SHAP enhances explanation diversity using nearby points.

ER-SHAP-RF leverages random forests for feature selection.

Abstract

Ensemble-based modifications of the well-known SHapley Additive exPlanations (SHAP) method for the local explanation of a black-box model are proposed. The modifications aim to simplify SHAP which is computationally expensive when there is a large number of features. The main idea behind the proposed modifications is to approximate SHAP by an ensemble of SHAPs with a smaller number of features. According to the first modification, called ER-SHAP, several features are randomly selected many times from the feature set, and Shapley values for the features are computed by means of "small" SHAPs. The explanation results are averaged to get the final Shapley values. According to the second modification, called ERW-SHAP, several points are generated around the explained instance for diversity purposes, and results of their explanation are combined with weights depending on distances between points and the explained instance. The third modification, called ER-SHAP-RF, uses the random forest for preliminary explanation of instances and determining a feature probability distribution which is applied to selection of features in the ensemble-based procedure of ER-SHAP. Many numerical experiments illustrating the proposed modifications demonstrate their efficiency and properties for local explanation.