Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem.

Matthias Hein, Maksym Andriushchenko, Julian Bitterwolf

Browse the full CVPR paper archive.