Upper Entropy for 2-Monotone Lower Probabilities

2026-03-23

Abstract

Uncertainty quantification is a key aspect in many tasks such as model selection/regularization, or quantifying prediction uncertainties to perform active learning or OOD detection. Within credal approaches that consider modeling uncertainty as probability sets, upper entropy plays a central role as an uncertainty measure. This paper is devoted to the computational aspect of upper entropies, providing an exhaustive algorithmic and complexity analysis of the problem. In particular, we show that the problem has a strongly polynomial solution, and propose many significant improvements over past algorithms proposed for 2-monotone lower probabilities and their specific cases.

Problem

• upper entropy
• 2-monotone lower probabilities
• uncertainty quantification
• computational complexity
• credal sets

Innovation

• upper entropy
• 2-monotone lower probabilities
• strongly polynomial algorithm
• uncertainty quantification
• credal sets