Machine Learning for Network Security

Change of Time: 12:00pm Eastern to accommodate our European time zone speaker!

With the wide application of Machine Learning (ML) for network security, new questions rise about the robustness of such models to prevent adversarial attacks. In this context, obfuscation techniques, such as mutation and morphing, present main challenges towards relying on ML-based IDSes. This talk will present the application of ML for network security in addition to introducing the existing obfuscation techniques along with possible (ML-based) solutions for detecting obfuscated attack traffic.

Ola Salman received the B.E. degree in Computer and Communications Engineering with distinction from the Lebanese University in 2013. In 2014, Ola joined the PhD accelerated track program in the Electrical and Computer Engineering (ECE) department at the American University of Beirut (AUB). After that, she joined the University of Helsinki, as a PostDoc researcher in the Machine Learning domain. Recently, she joined the DeepVu company as a Data Scientist. Her research interests are in Information Security and Networks, Software Defined Networks, Edge Computing, Artificial Intelligence, Internet of Things, and Machine Learning for cybersecurity. She had authored and co-authored more than 40 conference and journal articles.