Uncertainty-Aware Concept Bottleneck Models with Enhanced Interpretability
This work addresses interpretability and uncertainty in image classification for AI safety and reliability, representing an incremental improvement over existing CBMs.
The paper tackles the performance and uncertainty propagation issues in Concept Bottleneck Models (CBMs) for image classification by proposing a novel uncertainty-aware classifier that uses binary class-level concept prototypes for scoring and uncertainty measurement, achieving enhanced interpretability and robustness.
In the context of image classification, Concept Bottleneck Models (CBMs) first embed images into a set of human-understandable concepts, followed by an intrinsically interpretable classifier that predicts labels based on these intermediate representations. While CBMs offer a semantically meaningful and interpretable classification pipeline, they often sacrifice predictive performance compared to end-to-end convolutional neural networks. Moreover, the propagation of uncertainty from concept predictions to final label decisions remains underexplored. In this paper, we propose a novel uncertainty-aware and interpretable classifier for the second stage of CBMs. Our method learns a set of binary class-level concept prototypes and uses the distances between predicted concept vectors and each class prototype as both a classification score and a measure of uncertainty. These prototypes also serve as interpretable classification rules, indicating which concepts should be present in an image to justify a specific class prediction. The proposed framework enhances both interpretability and robustness by enabling conformal prediction for uncertain or outlier inputs based on their deviation from the learned binary class-level concept prototypes.