Improving the Weighting Strategy in KernelSHAP

TL;DR

This paper introduces a deterministic weighting modification to KernelSHAP, significantly reducing computation time and variance in Shapley value estimation for explainable AI, especially for high-dimensional data.

Contribution

The authors propose a novel deterministic weighting scheme for KernelSHAP that decreases variance and computational cost, enhancing efficiency in model explanation tasks.

Findings

Reduces the number of contribution function evaluations by up to 50%

Maintains accuracy of Shapley value approximations despite fewer evaluations

Enables explanation of higher-dimensional models within feasible runtimes

Abstract

In Explainable AI (XAI), Shapley values are a popular model-agnostic framework for explaining predictions made by complex machine learning models. The computation of Shapley values requires estimating non-trivial contribution functions representing predictions with only a subset of the features present. As the number of these terms grows exponentially with the number of features, computational costs escalate rapidly, creating a pressing need for efficient and accurate approximation methods. For tabular data, the KernelSHAP framework is considered the state-of-the-art model-agnostic approximation framework. KernelSHAP approximates the Shapley values using a weighted sample of the contribution functions for different feature subsets. We propose a novel modification of KernelSHAP which replaces the stochastic weights with deterministic ones to reduce the variance of the resulting Shapley value approximations. This may also be combined with our simple, yet effective modification to the KernelSHAP variant implemented in the popular Python library SHAP. Additionally, we provide an overview of established methods. Numerical experiments demonstrate that our methods can reduce the required number of contribution function evaluations by $5\%$ to $50\%$ while preserving the same accuracy of the approximated Shapley values -- essentially reducing the running time by up to $50\%$. These computational advancements push the boundaries of the feature dimensionality and number of predictions that can be accurately explained with Shapley values within a feasible runtime.