Towards a quantitative theory of adversarial cyber-risk analysis - Prof. Luca Allodi; 12/06/2018, 10:30 - 12:30



Towards a quantitative theory of adversarial cyber-risk analysis

Photo by Kevin on Unsplash

Prof. Luca Allodi, Assistant Professor

Affiliazione: Eindhoven University of Technology, Faculty of Mathematics and Computer Science; Security Group.

Expertise: Cybercrime and risk measurement

Durata: 2 ore
Programma: 12 Giugno 2018, ore 10:30 - 12:30
Aula Mocci - Padiglione A, DIEE -Dipartimento di Ingegneria Elettrica ed Elettronica, Università di Cagliari

In this talk we discuss steps towards a (more) realistic risk analysis framework for IT-systems, moving away from qualitative assumptions and risk estimates. We argue that in order to achieve this, we must adopt a more evidence-based adversarial perspective to: 1) devise more realistic attacker models; and 2) obtain empirical evidence of attacker decisions and attack deployment. This has to include aspects of an attack ranging from technical and economic incentives for malware production, to human factors behind a social engineering attack. 

To get there, in this talk we will evaluate data on real attacks at scale, explore the economics of a prominent underground (Russian) cybercrime market, and consider psychological aspects exploited by attackers behind a real social engineering campaign. Along the way, we will present metrics and models that can account for this effects, demonstrate them in action, and identify steps forward in this direction. The underlying thread running beneath the surface remains the same: can we reverse-engineer the attack process to meaningfully identify (likely) cyber-threats and measure the corresponding risk?


Davide Ariu, Battista Biggio, Matteo Mauri

Dip. di Ingegneria Elettrica ed Elettronica

Università di Cagliari