Alike Parts: A Feature-Informed Approach to Local and Global Prototype Explanations

TL;DR

This paper presents a feature-informed framework for prototype explanations that enhances interpretability by incorporating feature importance at local and global levels, improving explanation granularity and diversity.

Contribution

It introduces 'alike parts' for local explanations and a feature importance term in prototype selection, advancing prototype-based interpretability methods.

Findings

Maintains or increases surrogate model fidelity with feature-diverse prototypes.

Enhances local explanations by highlighting shared relevant feature subsets.

Promotes diversity in global prototype explanations without sacrificing accuracy.

Abstract

Prototype-based explanations offer an intuitive, example-based approach to support the interpretability of machine learning black box classifiers but often lack feature-level granularity. We introduce a framework that integrates feature importance at two levels to address this gap. First, for local explanations, we propose \textit{alike parts}: a method that uses feature importance scores to highlight the most relevant, shared feature subsets between a classified instance and its nearest prototype, guiding user attention. Second, we augment the global prototype selection objective function with a feature importance term to actively promote diversity in the feature attributions of the selected prototypes. Experiments on six benchmark datasets show that this augmented selection process maintains or, in some cases, increases the prediction fidelity of the surrogate model, suggesting that feature diversity does not compromise model fidelity.