70 citations · 230 across the 33 of their papers we have counts for
8 papers · 1 filter
On Fragile Features and Batch Normalization in Adversarial Training
Nils Philipp Walter, David Stutz, Bernt Schiele
Modern deep learning architecture utilize batch normalization (BN) to stabilize training and improve accuracy. It has been shown that the BN layers alone are surprisingly expressiv…
Relating Adversarially Robust Generalization to Flat Minima
David Stutz, Matthias Hein, Bernt Schiele
Adversarial training (AT) has become the de-facto standard to obtain models robust against adversarial examples. However, AT exhibits severe robust overfitting: cross-entropy loss…
Bit Error Robustness for Energy-Efficient DNN Accelerators
David Stutz, Nandhini Chandramoorthy, Matthias Hein +1
Deep neural network (DNN) accelerators received considerable attention in past years due to saved energy compared to mainstream hardware. Low-voltage operation of DNN accelerators…
Normalizing Flows with Multi-Scale Autoregressive Priors
Shweta Mahajan, Apratim Bhattacharyya, Mario Fritz +2
Flow-based generative models are an important class of exact inference models that admit efficient inference and sampling for image synthesis. Owing to the efficiency constraints o…
Confidence-Calibrated Adversarial Training: Generalizing to Unseen Attacks
David Stutz, Matthias Hein, Bernt Schiele
Adversarial training yields robust models against a specific threat model, e.g., adversarial examples. Typically robustness does not generalize to previously unseen thre…
"Best-of-Many-Samples" Distribution Matching
Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
Generative Adversarial Networks (GANs) can achieve state-of-the-art sample quality in generative modelling tasks but suffer from the mode collapse problem. Variational Autoencoders…