English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Open

Presents the pdf analysis as premium feature

PDF Analysis

Insights

Presents the summarisation as premium feature

Summarisation

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Nowadays, intrusion detection systems based on deep learning deliverstate-of-the-art performance. However, recent research has shown that speciallycrafted perturbations, called adversarial examples, are capable ofsignificantly reducing the performance of these intrusion detection systems.The objective of this paper is to design an efficient transfer learning-basedadversarial detector and then to assess the effectiveness of using multiplestrategically placed adversarial detectors compared to a single adversarialdetector for intrusion detection systems. In our experiments, we implementexisting state-of-the-art models for intrusion detection. We then attack thosemodels with a set of chosen evasion attacks. In an attempt to detect thoseadversarial attacks, we design and implement multiple transfer learning-basedadversarial detectors, each receiving a subset of the information passedthrough the IDS. By combining their respective decisions, we illustrate thatcombining multiple detectors can further improve the detectability ofadversarial traffic compared to a single detector in the case of a parallel IDSdesign.

TAD: Transfer Learning-based Multi-Adversarial Detection of Evasion Attacks against Network Intrusion Detection Systems