Wasim Essbai
Projektass. / MSc
Research Areas
- Machine Learning, Cybersecurity, Neural Networks, Cyber-Physical Systems
About
My research focuses on trustworthy AI, with an emphasis on runtime monitoring, anomaly detection, and verification of neural networks in safety-critical systems. I study methods to detect unreliable model behavior and assess the monitorability of learned representations, combining machine learning, statistical modeling, and formal verification.
Role
-
PreDoc Researcher
Cyber-Physical Systems, E191-01
Publications
-
A Comparison of Monitoring Techniques for Deep Neural Networks
/
Bartocci, E., & Essbai, W. (2024). A Comparison of Monitoring Techniques for Deep Neural Networks. In Bridging the Gap Between AI and Reality : Second International Conference, AISoLA 2024, Crete, Greece, October 30 – November 3, 2024, Proceedings. AISoLA 2024: International Conference on Bridging the Gap between AI and Reality, Kreta, Greece. https://doi.org/10.1007/978-3-031-75434-0_13
Projects: ProbInG (2020–2025) / TA-CPS (2023–2028) - A Framework for Including Uncertainty in Robustness Evaluation of Bayesian Neural Network Classifiers / Essbai, W., BOMBARDA, A., Bonfanti, S., & Gargantini, A. (2024). A Framework for Including Uncertainty in Robustness Evaluation of Bayesian Neural Network Classifiers. In DeepTest ’24: Proceedings of the 5th IEEE/ACM International Workshop on Deep Learning for Testing and Testing for Deep Learning (pp. 25–32). https://doi.org/10.1145/3643786.3648026