Anomaly detection in computer networks is a classic, long-term research problem. Previous attempts to solve it have used auto-encoders to learn a representation of the normal behaviour of networks and detect anomalies according to reconstruction loss. In this talk, I'll present the concept of "auto-encoder losses transfer learning". This approach normalizes auto-encoder losses in different model deployments, providing the ability to transform loss vectors of different networks with potentially significant varying characteristics, properties, and behaviors into a domain invariant representation. This is forwarded to a global detection model that can detect and classify threats in a generalized way that is agnostic to the specific network deployment, allowing for comprehensive network coverage. This talk is based on a recently presented paper in ACM CCS AISec 21'.
Enthusiastic about math and computer science from childhood, Dr. Yehezkel began his academic path towards a BSc in applied mathematics during high school (summa cum laude). After 5 years of IDF service as a Senior Research Captain in Naval Intelligence and unit 8200, Aviv earned his Computer Science PhD from the Technion, his PhD thesis was about novel computer networks and cybersecurity big data problems and applications.
After spending a year with NICE systems as senior researcher, Dr. Yehezkel joined Eyal Elyashiv to co-found Cynamics as the company CTO, leading innovation, technology and R&D.