AI enhances network intrusion detection, also against AI adversaries!
Network intrusion detection systems (NIDS) are central in securing critical infrastructure, as cyberattacks rely on penetrating industrial control system networks. Artificial Intelligence (AI) techniques are increasingly used in detecting malicious network traffic. However, at the same time, AI can also come to the aid of attackers and offer new avenues for attacks.
To address this challenge, new techniques based on so-called differentiable logics have been proposed to train AI models that are robust against adversarial AI attacks.
In this workshop, we will present a machine learning model for network intrusion detection developed at ITU and based on differentiable logics. This novel technique allows training classifiers that are more robust, which in turn helps the model generalise more and increase its performance in real world scenarios, including detecting adversarial AI attacks.
The workshop is mainly targeted to companies using network intrusion detection. Basic understanding of machine learning is advisable.
Zaruhi Aslanyan
Senior Security Architect, PhD
Alexandra Instituttet
zaruhi.aslanyan@alexandra.dk
Måske kunne disse arrangementer også have din interesse:
12. JUNI 2025 · ONLINE
Bliv inspireret til at arbejde strategisk med digitalisering og bæredygtighed, så I kan træffe kloge valg. I webinaret møder du Director of Digital Sustainability, Trine Plambech.
12. JUNI 2025 · KØBENHAVN
Vi er med på AIDC-konferencen ‘Data as the Digital Heartbeat: Transparency and Value Creation in the Modern Value Chain. Mød bl.a. Lea Schick og Alexandre Alapetite fra Alexandra Instituttet.
17. JUNI 2025 · ONLINE
Deltag i vores 30 minutters webinar og lær, hvordan I undgår de værste faldgruber og kommer godt i mål, når I skal implementere ny teknologi.
Ⓒ 2025 - Alexandra Instituttet A/S
CVR nr. 24 21 33 66
