Credal Concept Bottleneck Models for Epistemic-Aleatoric Uncertainty Decomposition

TL;DR

This paper introduces CREDENCE, a framework for concept bottleneck models that decomposes uncertainty into epistemic and aleatoric components using credal predictions, enhancing interpretability and decision-making.

Contribution

CREDENCE is the first CBM framework to explicitly separate epistemic and aleatoric uncertainties at the concept level using credal predictions.

Findings

Epistemic uncertainty correlates with prediction errors.

Aleatoric uncertainty aligns with annotator disagreement.

CREDENCE improves decision-making by uncertainty decomposition.

Abstract

Concept Bottleneck Models (CBMs) predict through human-interpretable concepts, but they typically output point concept probabilities that conflate epistemic uncertainty (reducible model underspecification) with aleatoric uncertainty (irreducible input ambiguity). This makes concept-level uncertainty hard to interpret and, more importantly, hard to act upon. We introduce CREDENCE (Credal Ensemble Concept Estimation), a CBM framework that decomposes concept uncertainty by construction. CREDENCE represents each concept as a credal prediction (a probability interval), derives epistemic uncertainty from disagreement across diverse concept heads, and estimates aleatoric uncertainty via a dedicated ambiguity output trained to match annotator disagreement when available. The resulting signals support prescriptive decisions: automate low-uncertainty cases, prioritize data collection for high-epistemic cases, route high-aleatoric cases to human review, and abstain when both are high. Across several tasks, we show that epistemic uncertainty is positively associated with prediction errors, whereas aleatoric uncertainty closely tracks annotator disagreement, providing guidance beyond error correlation. Our implementation is available at the following link: https://github.com/Tankiit/Credal_Sets/tree/ensemble-credal-cbm