Hussein Fawaz
Ph.D. Student
PhD Student & Software Engineer Trustworthy AI for Cybersecurity
HFAbout
Hi, I'm Hussein 👋 — a Ph.D. student in Informatics at Università della Svizzera italiana (USI) and the University of Applied Sciences and Arts of Southern Switzerland (SUPSI) in Lugano, Switzerland. I am part of the Institute of Information Systems and Networking (ISIN) at SUPSI.
I am supervised by Prof. Marc Langheinrich (USI), Prof. Silvia Giordano (SUPSI), and Dr. Omran Ayoub (SUPSI).
Before starting my Ph.D., I earned my M.Sc. in Information Systems and Data Intelligence as well as a B.Sc. in Computer Science from the Lebanese University. Alongside research, I have several years of experience as a software engineer, building secure and scalable full-stack systems.
Research Interests: Trustworthy AI, Explainable AI, Network Intrusion Detection, Uncertainty Quantification.
Latest News
Teaching Assistant (USI)
Teaching Assistant (USI)
🎓 TA for Business Process Modeling, Management and Mining (Prof. Cesare Pautasso, USI Spring 2026)
Paper Accepted (WONS 2026)
Paper Accepted (WONS 2026)
📄 H. Fawaz, O. Ayoub, D. Andreoletti, S. Giordano — "Energy Cost of Enhancing Reliability of Machine Learning Models for Edge IoT Security". IEEE WONS 2026.
Paper Published (WiMob 2025)
Paper Published (WiMob 2025)
📄 H. Fawaz, F. Ezzeddine, S. Giordano, O. Ayoub — "Towards Better-Calibrated ML Models for Reliable Network Intrusion Detection via Calibration-Aware SHAP-Based Feature Selection". WiMob 2025, Marrakesh, Morocco.
Teaching Assistant (SUPSI)
Teaching Assistant (SUPSI)
🎓 TA for Algorithmic Design (Dr. Omran Ayoub, SUPSI Fall 2025).
Web Chair
Web Chair
🧩 Web Chair for IFIP Networking 2026 (NETWORKING 2026) and TX4Nets 2025 (2nd International Workshop on Trustworthy and eXplainable AI for Networks).
Publications
View the full list of my publications on
Energy Cost of Enhancing Reliability of Machine Learning Models for Edge IoT Security
Energy Cost of Enhancing Reliability of Machine Learning Models for Edge IoT Security
Study on the energy cost of improving reliability of ML models for edge IoT security.
Authors: H. Fawaz, O. Ayoub, D. Andreoletti, S. Giordano
Towards Better-Calibrated ML Models for Reliable Network Intrusion Detection via Calibration-Aware SHAP-Based Feature Selection
Towards Better-Calibrated ML Models for Reliable Network Intrusion Detection via Calibration-Aware SHAP-Based Feature Selection
Calibration-aware SHAP-based feature selection to improve reliability and calibration for network intrusion detection models.
Authors: H. Fawaz, F. Ezzeddine, S. Giordano, O. Ayoub
Reducing Complexity and Enhancing Predictive Power of ML-based Lightpath Quality of Transmission Estimation via SHAP-Assisted Feature Selection
Reducing Complexity and Enhancing Predictive Power of ML-based Lightpath Quality of Transmission Estimation via SHAP-Assisted Feature Selection
SHAP-assisted feature selection to reduce complexity and improve ML-based lightpath QoT estimation in optical networks.
Authors: H. Fawaz, F. Arpanaei, D. Andreoletti, I. Sbeity, J. A. Hernández, D. Larrabeiti, O. Ayoub
Education
Università della Svizzera italiana (USI) - SUPSI
Lebanese University
Lebanese University
Work Experience
SUPSI
AGParts
CloudGate Consulting DWC-LLC
Lebanese International University (LIU)
Academic Services
Teaching Assistant: Business Process Modeling, Management and Mining (Prof. Cesare Pautasso)
Teaching Assistant: Algorithmic Design (Dr. Omran Ayoub)
Awards & Honors
Swiss Government Excellence Scholarship (ESKAS)
Skills
Check out my latest work
Trustworthy ML for Network Intrusion Detection (Ph.D. Research)
Trustworthy ML for Network Intrusion Detection (Ph.D. Research)
Research on trustworthy and reliable machine learning for cybersecurity applications, with a focus on network intrusion detection systems (NIDS), explainability, privacy-preserving learning, and uncertainty quantification.
Calibration-Aware SHAP-Based Feature Selection for Reliable NIDS
Calibration-Aware SHAP-Based Feature Selection for Reliable NIDS
Feature selection using SHAP with calibration-aware criteria to improve reliability of ML models for network intrusion detection.
Get in Touch
Want to chat? Feel free to reach out via email →
- • Explore collaboration opportunities