On the Interpretability of Part-Prototype Based Classifiers: A Human Centric Analysis

TL;DR

This paper introduces a human-centric framework with actionable metrics to evaluate the interpretability of part-prototype classifiers, validated through extensive Amazon Mechanical Turk experiments.

Contribution

It presents the first comprehensive, human-focused evaluation framework for part-prototype models, including three metrics and experimental validation.

Findings

Framework effectively assesses interpretability of part-prototype models

Experiments demonstrate the framework's capability across various models

Most comprehensive evaluation of part-prototype interpretability to date

Abstract

Part-prototype networks have recently become methods of interest as an interpretable alternative to many of the current black-box image classifiers. However, the interpretability of these methods from the perspective of human users has not been sufficiently explored. In this work, we have devised a framework for evaluating the interpretability of part-prototype-based models from a human perspective. The proposed framework consists of three actionable metrics and experiments. To demonstrate the usefulness of our framework, we performed an extensive set of experiments using Amazon Mechanical Turk. They not only show the capability of our framework in assessing the interpretability of various part-prototype-based models, but they also are, to the best of our knowledge, the most comprehensive work on evaluating such methods in a unified framework.