Me in a Nutshell

I am a final-year PhD candidate in the Department of Computing at Imperial College London, supervised by Dr. Sergio Maffeis. My research focuses on improving deep learning-based network intrusion detection systems (NIDS). Specifically, I am trying to:

• Adapt one-class anomaly detection NIDS to handle changes in normal (benign) behavior.
• Enable multi-class NIDS to adapt to new attacks as they emerge over time.
• Enhance the robustness of these systems against noisy labels and label-flipping attacks.

By adapt, I mean developing systems that can detect and adjust to changes with minimal cost.

Updates

• [2024-11] Euan Goodbrand's thesis won The Corporate Partnership Programme Individual Project Prize in Computing Science.
• [2024-11] Euan Goodbrand's thesis earned the designation of Distinguished Thesis.
• [2024-7] YuNan Peng's thesis was awarded a Distinction.
• [2024-6] Mateen is accepted at RAID 2024
• [2024-4] Rasd is accepted at IFIP SEC 2024

Publications

Mateen: Adaptive Ensemble Learning for Network Anomaly Detection with Sergio Maffeis. Appeared in RAID 2024.

Rasd: Semantic Shift Detection and Adaptation for Network Intrusion with Sergio Maffeis. Appeared in IFIP SEC 2024.

Supervision

I co-supervised the following students alongside Dr. Sergio Maffeis.

• [2024] Euan Goodbrand - MSc degree in Artificial Intelligence and Machine Learning with Distinction from ICL.
Project: Deep Learning From Noisy and Imbalanced Security Data
• [2024] YuNan Peng - MEng degree in Electronic and Information Engineering with Distinction from ICL.
Project: Enhancing Network Intrusion Detection Systems with Active Learning: A Semi-Supervised Coreset Selection Approach

Review

I co-reviewed for the following conferences alongside Dr. Sergio Maffeis.

• [2024] IEEE Computer Security Foundations Symposium (CSF)
• [2023] IEEE Security and Privacy (Oakland)
• [2021] IEEE Security and Privacy (Oakland)