We are also working on novel ways to collect the massive datasets necessary to train these algorithms. Here we should acknowledge the tireless dedication of hackers, particularly in Russia and China.
The application of probabilistic methods are long overdue in the field of information security. Although research in this area has been extremely limited, initial experiments appear promising.
Doupé, Adam, Marco Cova, and Giovanni Vigna. Why Johnny can’t pentest: An analysis of black-box web vulnerabilit…. International Conference on Detection of Intrusions and Malware, and Vulnerability Assessment. Springer Berlin Heidelberg, 2010.
Sarraute, Carlos, Olivier Buffet, and Jörg Hoffmann. Penetration Testing== POMDP Solving? arXiv preprint arXiv:1306.4714 (2013).
Link: Probabilistic Pentesting