While deep neural networks have been very successfully employed in classification problems, their stability properties remain still unclear. In particular, the presence of so-called adversarial examples has demonstrated that state-of-the-art networks are extremely vulnerable to small perturbations in the data.
Lernziel
In this seminar, we will consider the state-of-the-art in adversarial attacks and defenses.
Voraussetzungen / Besonderes
Participants should already be familiar with the principles of deep neural networks. The course will also include programming that will require knowledge in using either PyTorch or Tensorflow.
Leistungskontrolle
Information zur Leistungskontrolle (gültig bis die Lerneinheit neu gelesen wird)
Leistungskontrolle als Semesterkurs