Adversarial examples are among the most widespread attacks in adversarial machine learning. In this work, we define new targeted and non-targeted attacks that are computationally less expensive than standard adversarial attacks. Besides practical purposes in some scenarios, these attacks can improve our understanding of the robustness of machine learning models. Moreover, we introduce a new training scheme to improve the performance of pre-trained neural networks and defend against our attacks. We examine the differences between our method, standard training, and standard adversarial training on pre-trained models. We find that our method protects the networks better against our attacks. Furthermore, unlike usual adversarial training, which reduces standard accuracy when applied to previously trained networks, our method maintains and sometimes even improves standard accuracy.
Identifer | oai:union.ndltd.org:kaust.edu.sa/oai:repository.kaust.edu.sa:10754/678525 |
Date | 06 1900 |
Creators | Fourati, Fares |
Contributors | Alouini, Mohamed-Slim, Computer, Electrical and Mathematical Science and Engineering (CEMSE) Division, Kammoun, Abla, Gao, Xin, Elatab, Nazek |
Source Sets | King Abdullah University of Science and Technology |
Language | English |
Detected Language | English |
Type | Thesis |
Rights | 2023-06-02, At the time of archiving, the student author of this thesis opted to temporarily restrict access to it. The full text of this thesis will become available to the public after the expiration of the embargo on 2023-06-02. |
Page generated in 0.0125 seconds