The Weighting Game: Evaluating Quality of Explainability Methods
Abstract
The objective of this paper is to assess the quality of explanation heatmaps for image classification tasks. To assess the quality of explainability methods, we approach the task through the lens of accuracy and stability. In this work, we make the following contributions. Firstly, we introduce the Weighting Game, which measures how much of a class-guided explanation is contained within the correct class' segmentation mask. Secondly, we introduce a metric for explanation stability, using zooming/panning transformations to measure differences between saliency maps with similar contents. Quantitative experiments are produced, using these new metrics, to evaluate the quality of explanations provided by commonly used CAM methods. The quality of explanations is also contrasted between different model architectures, with findings highlighting the need to consider model architecture when choosing an explainability method.
Cite
@article{arxiv.2208.06175,
title = {The Weighting Game: Evaluating Quality of Explainability Methods},
author = {Lassi Raatikainen and Esa Rahtu},
journal= {arXiv preprint arXiv:2208.06175},
year = {2025}
}
Comments
Published in: Image Analysis (SCIA 2025), Lecture Notes in Computer Science (LNCS), vol. 15726, pp. 325-338 (2025). This is the submitted-manuscript (pre-review) version. v2: added required preprint notice and updated metadata. Version of Record: see DOI 10.1007/978-3-031-95918-9_23