Finding Patterns in Ambiguity: Interpretable Stress Testing in the Decision Boundary

Zenodo (CERN European Organization for Nuclear Research)

Abstract

The increasing use of deep learning across various domains highlights the importance of understanding the decision-making processes of these black-box models. Recent research focusing on the decision boundaries of deep classifiers relies on generated synthetic instances in areas of low confidence uncovering samples that challenge both models and humans. We propose a novel approach to enhance the interpretability of deep binary classifiers by selecting representative samples from the decision boundary - prototypes - and applying post-model explanation algorithms. We evaluate the effectiveness of our approach through 2D visualizations and GradientSHAP analysis. Our experiments demonstrate the potential of the proposed method revealing distinct and compact clusters and diverse prototypes that capture essential features that lead to low-confidence decisions. By offering a more aggregated view of deep classifiers' decision boundaries our work contributes to the responsible development and deployment of reliable machine learning systems.

Authors 6

1. I. L. V. GOMES

   Universidade do Porto

   Affiliation as printed

   Faculty of Engineering - University of Porto (FEUP)

2. Luís Filipe Teixeira

   Universidade do Porto

   Affiliation as printed

   Faculty of Engineering - University of Porto (FEUP)

3. Jan N. van Rijn Aachen

   Leiden University

   Affiliation as printed

   Leiden University

4. Carlos Soares

   Universidade do Porto · Fraunhofer Portugal Research

   Affiliation as printed

   Faculty of Engineering - University of Porto (FEUP)

   Artificial Intelligence and Computer Science Laboratory (LIACC)

   Fraunhofer Portugal Research

5. André Restivo

   Universidade do Porto

   Affiliation as printed

   Faculty of Engineering - University of Porto (FEUP)

6. Moisés Santos

   Universidade do Porto

   Affiliation as printed

   Faculty of Engineering - University of Porto (FEUP)