Rima Alaifari: Catalogue data in Autumn Semester 2020
|Prof. Dr. Rima Alaifari
Seminar für Angewandte Mathematik
ETH Zürich, HG G 59.2
|+41 44 632 32 00
|Robustness of Deep Neural Networks
Number of participants limited to 40
|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.
|In this seminar, we will consider the state-of-the-art in adversarial attacks and defenses.
|Prerequisites / Notice
|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.
|Zurich Colloquium in Applied and Computational Mathematics
|R. Abgrall, R. Alaifari, H. Ammari, R. Hiptmair, S. Mishra, S. Sauter, C. Schwab