Why Uncertainty Calibration Matters for Reliable Perturbation-based Explanations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Decker, Thomas, Tresp, Volker, Buettner, Florian
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911020510871552
author Decker, Thomas
Tresp, Volker
Buettner, Florian
author_facet Decker, Thomas
Tresp, Volker
Buettner, Florian
contents Perturbation-based explanations are widely utilized to enhance the transparency of modern machine-learning models. However, their reliability is often compromised by the unknown model behavior under the specific perturbations used. This paper investigates the relationship between uncertainty calibration - the alignment of model confidence with actual accuracy - and perturbation-based explanations. We show that models frequently produce unreliable probability estimates when subjected to explainability-specific perturbations and theoretically prove that this directly undermines explanation quality. To address this, we introduce ReCalX, a novel approach to recalibrate models for improved perturbation-based explanations while preserving their original predictions. Experiments on popular computer vision models demonstrate that our calibration strategy produces explanations that are more aligned with human perception and actual object locations.
format Preprint
id arxiv_https___arxiv_org_abs_2506_19630
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Why Uncertainty Calibration Matters for Reliable Perturbation-based Explanations
Decker, Thomas
Tresp, Volker
Buettner, Florian
Machine Learning
Artificial Intelligence
Perturbation-based explanations are widely utilized to enhance the transparency of modern machine-learning models. However, their reliability is often compromised by the unknown model behavior under the specific perturbations used. This paper investigates the relationship between uncertainty calibration - the alignment of model confidence with actual accuracy - and perturbation-based explanations. We show that models frequently produce unreliable probability estimates when subjected to explainability-specific perturbations and theoretically prove that this directly undermines explanation quality. To address this, we introduce ReCalX, a novel approach to recalibrate models for improved perturbation-based explanations while preserving their original predictions. Experiments on popular computer vision models demonstrate that our calibration strategy produces explanations that are more aligned with human perception and actual object locations.
title Why Uncertainty Calibration Matters for Reliable Perturbation-based Explanations
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2506.19630