Classification Metrics for Image Explanations: Towards Building Reliable XAI-Evaluations

TL;DR

This paper introduces new metrics for evaluating saliency explanations in image classification, benchmarks existing methods on ImageNet, and proposes a reliability scheme inspired by psychometric testing to improve XAI evaluation.

Contribution

It develops novel evaluation metrics for saliency methods and proposes a reliability assessment scheme, addressing the limitations of current proxy metrics in XAI.

Findings

New metrics for saliency explanation evaluation

Benchmarking of common saliency methods on ImageNet

A reliability evaluation scheme for explanation metrics

Abstract

Decision processes of computer vision models - especially deep neural networks - are opaque in nature, meaning that these decisions cannot be understood by humans. Thus, over the last years, many methods to provide human-understandable explanations have been proposed. For image classification, the most common group are saliency methods, which provide (super-)pixelwise feature attribution scores for input images. But their evaluation still poses a problem, as their results cannot be simply compared to the unknown ground truth. To overcome this, a slew of different proxy metrics have been defined, which are - as the explainability methods themselves - often built on intuition and thus, are possibly unreliable. In this paper, new evaluation metrics for saliency methods are developed and common saliency methods are benchmarked on ImageNet. In addition, a scheme for reliability evaluation of such metrics is proposed that is based on concepts from psychometric testing. The used code can be found at https://github.com/lelo204/ClassificationMetricsForImageExplanations .