2 papers
cs.LG2024
On the Calibration of Epistemic Uncertainty: Principles, Paradoxes and Conflictual Loss
Mohammed Fellaji, Frédéric Pennerath, Brieuc Conan-Guez +1
The calibration of predictive distributions has been widely studied in deep learning, but the same cannot be said about the more specific epistemic uncertainty as produced by Deep…
stat.ML2024
The Epistemic Uncertainty Hole: an issue of Bayesian Neural Networks
Mohammed Fellaji, Frédéric Pennerath
Bayesian Deep Learning (BDL) gives access not only to aleatoric uncertainty, as standard neural networks already do, but also to epistemic uncertainty, a measure of confidence a mo…