no code implementations • 18 Aug 2023 • Johan Jonasson, Jeffrey E. Steif, Olof Zetterqvist
A first step is taken towards understanding often observed non-robustness phenomena of deep neural net (DNN) classifiers.
no code implementations • 15 Dec 2021 • Olof Zetterqvist, Rebecka Jörnsten, Johan Jonasson
Here we propose a novel double regularization of the neural network training loss that combines a penalty on the complexity of the classification model and an optimal reweighting of training observations.