Kontakt

28 January 2026

Training AI to improve detection of cyber threats

Discover how neural networks can be trained with logical rules to detect malicious network traffic while reducing false alerts. In this workshop, we will show how differentiable logics can boost both performance and explainability, with applications in cybersecurity and beyond.
  • 28 January 2026
  • 09:00 – 11:30
  • IT University of Denmark
    Rued Langgaards Vej 7, 2300 Copenhagen
    Room TBA
  • Language: English
  • Participation is free
  • Organizers: The Alexandra Institute in collaboration with ITU and DTU
  • Registration deadline: 26 January 2026 at 13:00
Sign up here

about the workshop

Robust AI for network security

Network intrusion detection systems (NIDS) are essential for securing critical infrastructure, as cyberattacks often target industrial control system networks. AI techniques are increasingly used to detect malicious network traffic – but AI can also be exploited by attackers. To address this challenge, new techniques based on differentiable logics have been proposed to train AI models that are robust against adversarial AI attacks.

As part of an NFC project, we have developed a framework for training neural networks to satisfy logical constraints using  differentiable logics. This framework enabled a classifier that can detect malicious network traffic, that is robust against adversarial attacks and includes human domain knowledge. We aim to reduce alert fatigue by ruling out clearly benign network traffic while capturing typical attacks like DoS.

The classifier performance is consistent across different datasets, allowing NIDS to react to changes in the traffic distribution. In this workshop, we will demonstrate how differentiable logics enhance both prediction performance and explainability.

Target group

Who can participate?

The workshop is mainly targeted to companies using network intrusion detection. Basic understanding of machine learning is advisable. Everybody is welcome, though the workshop is intended for a technical audience.

Workshop agenda

09:00-09:15 Introduction, presentation of the project and partners
09:15-10:00 Presentation of results
10:00-10:15 Break
10:15-11:15 Discussion for future directions and experimental integration into industrial case studies
11:15-11:30 Conclusion and farewell

Benefits

Key takeaways

  • Understand how to combine data-driven learning with logical specifications using differentiable logics to enhance AI models for cybersecurity.

  • See practical demonstrations of this approach in Network Intrusion Detection Systems (NIDS), showcasing its real-world applicability.

  • Learn how these techniques improve robustness against adversarial attacks, resulting in more secure and reliable intrusion detection solutions.

Contact

Do you want to know more?

If you want to know more, please contact:

Zaruhi Aslanyan
Senior Security Architect, PhD
Alexandra Institute

+45 93 50 87 40
zaruhi.aslanyan@alexandra.dk

Andre events

Måske kunne disse events også have din interesse:
Arrangement

Danmarks kritiske infrastruktur

8. SEPTEMBER 2026 · KASTRUP
En kurateret konference om Danmarks kritiske infrastruktur. Vi undersøger, hvor vi står i dag – og hvad der skal til for at styrke Danmarks robusthed.

Læs mere »
Arrangement

AI-konferencen 2026: Hvad betyder AI for fremtidens maskinmestre?

10. SEPTEMBER 2026 · FREDERICIA
Hvordan bruger virksomheder reelt data, automatisering, AI og robotter i deres daglige drift – og hvad virker faktisk? Mød virksomheder og eksperter, der deler konkrete erfaringer, dokumenterede resultater og de vigtigste læringer – også når implementeringen har været udfordrende.

Læs mere »

Formular indsendt!