Robustness of machine learning systems: an overview of Defence against Adversarial AI Attacks
other
Alan Turing once said: “A computer would deserve to be called intelligent if it could deceive a human into believing that it was human”. Currently, we cannot confirm that any system successfully deceived a human into believing that it is a human. However, there are plenty of cases of computers deceiving other computers, for example by fooling it into thinking a picture of a hamster is actually a burrito (Anley, 2022). The ability to deceive Artificial Intelligence (AI) models has sparked discussion among researchers about their robustness and safety. In order to counter the risks that come with Adversarial AI attacks, a novel study branch has emerged that deals with defence methods against such attacks.
TNO Identifier
1000779
Publisher
TNO
Collation
22 p.
Place of publication
Den Haag