Towards Desiderata-Driven Design of Visual Counterfactual Explainers
By: Sidney Bender , Jan Herrmann , Klaus-Robert Müller and more
Potential Business Impact:
Shows how to change pictures to fool computers.
Visual counterfactual explainers (VCEs) are a straightforward and promising approach to enhancing the transparency of image classifiers. VCEs complement other types of explanations, such as feature attribution, by revealing the specific data transformations to which a machine learning model responds most strongly. In this paper, we argue that existing VCEs focus too narrowly on optimizing sample quality or change minimality; they fail to consider the more holistic desiderata for an explanation, such as fidelity, understandability, and sufficiency. To address this shortcoming, we explore new mechanisms for counterfactual generation and investigate how they can help fulfill these desiderata. We combine these mechanisms into a novel 'smooth counterfactual explorer' (SCE) algorithm and demonstrate its effectiveness through systematic evaluations on synthetic and real data.
Similar Papers
Faithful Counterfactual Visual Explanations (FCVE)
CV and Pattern Recognition
Shows how computers "see" to make them smarter.
V-CECE: Visual Counterfactual Explanations via Conceptual Edits
CV and Pattern Recognition
Changes pictures to fool AI, showing how it thinks.
DocVCE: Diffusion-based Visual Counterfactual Explanations for Document Image Classification
CV and Pattern Recognition
Shows why computers decide documents are what they are.