Doktorandské kolokvium KAI - Iveta Bečková (6.3.2023)

v pondelok 6.3.2023 o 13:10 hod. v miestnosti I/9


01. 03. 2023 10.45 hod.
Od: Damas Gruska

Prednášajúci: Iveta Bečková

Názov: Adversarial Examples in Deep Learning

Termín: 6.3.2023, 13:10 hod., I/9


Abstrakt:
Deep neural networks achieve remarkable performance in multiple fields. However, after proper training they suffer from an inherent vulnerability against adversarial examples (AEs).The AEs attempt to find the worst-case perturbation in input space, resulting in faulty output (such as misclassification). Different methods of attacks provide different approximations of this worst-case and each of them has certain advantages and disadvantages.The problem gets even more complicated in the deep RL setting, where time is also a factor. We will present our work on comparing different adversarial attacks, as well as plans for future research in adversarial attacks on deep RL agents.