SHAP values via sparse Fourier representation

Ali Gorji, Andisheh Amrollahi, Andreas Krause

Keywords: SHAP, Interpretability, XAI, Fourier

TL;DR: Computing SHAP values for black-boxes and trees using sparse Fourier representation of models, orders of magnitudes faster than other methods

Abstract: SHAP (SHapley Additive exPlanations) values are a widely used method for local feature attribution in interpretable and explainable AI. We propose an efficient two-stage algorithm for computing SHAP values in both black-box setting and tree-based models. Motivated by spectral bias in real-world predictors, we first approximate models using compact Fourier representations, exactly for trees and approximately for black-box models. In the second stage, we introduce a closed-form formula for {\em exactly} computing SHAP values using the Fourier representation, that ``linearizes'' the computation into a simple summation and is amenable to parallelization. As the Fourier approximation is computed only once, our method enables amortized SHAP value computation, achieving significant speedups over existing methods and a tunable trade-off between efficiency and precision.

Primary Area: Social and economic aspects of machine learning (e.g., fairness, interpretability, human-AI interaction, privacy, safety, strategic behavior)

Submission Number: 23391