Improving uncertainty estimation in deep neural networks by modeling class and instance uncertainty

Dimitrios Spanos, Nikolaos Passalis, Anastasios Tefas

Abstract: Highlights•A deep learning method that models data uncertainty for uncertainty estimation is proposed.•A method that generates soft labels in an adaptive manner is used for the optimization.•The uncertainty-aware learning process improves state-of-the-art methodologies in out-of-distribution detection.

External IDs:dblp:journals/pr/SpanosPT26