Faculty of Mathematics, Physics
and Informatics
Comenius University Bratislava

Doctoral colloquium - Iveta Bečková (6.3.2023)

Monday 6.3.2023 at 13:10, Lecture room I/9

01. 03. 2023 10.57 hod.
By: Damas Gruska

Iveta Bečková:
Adversarial Examples in Deep Learning

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.