Conceptualizing Uncertainty: A Concept-based Approach to Explaining Uncertainty
This addresses the problem of limited interpretability in uncertainty quantification for high-dimensional data, offering a method to improve model refinement, though it appears incremental as it builds on existing concept activation techniques.
The paper tackles the challenge of explaining uncertainty in high-dimensional machine learning models by proposing a concept-based approach using concept activation vectors, which provides both local and global explanations to enhance interpretability and trust.
Uncertainty in machine learning refers to the degree of confidence or lack thereof in a model's predictions. While uncertainty quantification methods exist, explanations of uncertainty, especially in high-dimensional settings, remain an open challenge. Existing work focuses on feature attribution approaches which are restricted to local explanations. Understanding uncertainty, its origins, and characteristics on a global scale is crucial for enhancing interpretability and trust in a model's predictions. In this work, we propose to explain the uncertainty in high-dimensional data classification settings by means of concept activation vectors which give rise to local and global explanations of uncertainty. We demonstrate the utility of the generated explanations by leveraging them to refine and improve our model.