S-LIME: Stabilized-LIME for Model Explanation

TL;DR

S-LIME introduces a statistically grounded approach to stabilize local explanations of black-box models, improving reliability and user trust in high-stakes domains.

Contribution

It proposes a hypothesis testing framework based on the central limit theorem to determine the number of perturbations needed for stable explanations.

Findings

S-LIME reduces explanation instability in experiments.

The method guarantees explanation stability with fewer perturbations.

Demonstrates effectiveness on both simulated and real datasets.

Abstract

An increasing number of machine learning models have been deployed in domains with high stakes such as finance and healthcare. Despite their superior performances, many models are black boxes in nature which are hard to explain. There are growing efforts for researchers to develop methods to interpret these black-box models. Post hoc explanations based on perturbations, such as LIME, are widely used approaches to interpret a machine learning model after it has been built. This class of methods has been shown to exhibit large instability, posing serious challenges to the effectiveness of the method itself and harming user trust. In this paper, we propose S-LIME, which utilizes a hypothesis testing framework based on central limit theorem for determining the number of perturbation points needed to guarantee stability of the resulting explanation. Experiments on both simulated and real world data sets are provided to demonstrate the effectiveness of our method.