← Abdullah Ahmad Khan
Email
CURRICULUM VITAE · 2026

Abdullah Ahmad Khan

Perth, Western Australia
aahmad607@gmail.com
github.com/abdullahak07

Academic Appointments

Academic Lecturer

University of Notre Dame Australia · Perth

Lectures, tutorials and practical sessions in COMP6012 Cyber Threat Intelligence and COMP6013 Network Security & Cryptography, including assessment, marking and consultation.

Casual Academic · Information Technology

Murdoch University · Perth

Teaching across ICT619 Artificial Intelligence, ICT206 Intelligent Systems and ICT100 Transition to IT for undergraduate and postgraduate students.

Research Experience

PhD Researcher · Trustworthy AI

Murdoch University

  • Machine learning, multimodal AI, machine unlearning, privacy-preserving ML and VLM evaluation.
  • Reproducible pipelines for training, attack simulation, robustness tests, ablations and statistical evaluation.
  • Research into membership inference, quantisation-induced recovery, metric reliability and post-deployment model behaviour.
  • Lead first-author projects from method design through implementation, experiments, analysis, writing and artifact release.

Education

Doctor of Philosophy · Computer Science

Murdoch University

Thesis finalisation in machine unlearning and trustworthy multimodal AI.

Master of Computer Science & Applications

Aligarh Muslim University

Master’s Programme — Admission

University of Turku · Finland

Admitted to a master’s programme.

Selected Publications & Research Outputs

Neurocomputing · 2026Hessian-Guided Gradient Unlearning
Neural Networks · RevisionSelective Multimodal Unlearning via Boundary-Based Forgetting
Knowledge-Based Systems · Under reviewMU-ALIGN: Tail-Suppressed Multimodal Machine Unlearning
arXiv · 2026DurableUn: Quantization-Induced Recovery Attacks in Machine Unlearning
arXiv · 2026Metric Unreliability in Multimodal Machine Unlearning
Springer · 2023Rapid Face Mask Detection and Person Identification Model Based on Deep Neural Networks
IEEE · 2022Islamophobic Tweet Detection using Transfer Learning