Teaching

Courses I have taught and supervised (selected). Slides and materials are linked when available.

Deep Learning (Postgraduate) — Fall 2025 — ITU Istanbul

Main Lecturer 12 weeks • lectures + coding + project (PyTorch)

A 12-week postgraduate course covering fundamentals and modern deep learning architectures with emphasis on theory, implementation, and research discussions. Includes weekly lectures, coding sessions, and a final project.

Syllabus & materials (click to collapse)
  1. Week 1 Introduction to Deep Learning
  2. Week 2 Unsupervised Feature Learning as a Gateway to CNNs
  3. Week 3 Convolutional Neural Networks (CNNs)
  4. Week 4 Sequence Modeling — RNNs and LSTMs
  5. Week 5 Transformers
  6. Week 6 Large Language Models, MoE, DeepSeek
  7. Week 7 Generative Modeling (AE, VAE, GANs)
  8. Week 8 Generative Modeling — Autoregressive Models
  9. Week 9 Normalizing Flows and SDE
  10. Week 10 Self-Supervised Learning
  11. Week 11 Research Topics, Applications & Final Project Presentations
  12. Week 12 Final Project Presentations

Deep Learning (Postgraduate) — 2019–2021 — IPM

Main Lecturer 16 weeks • 2×1.5h/week

Covered classical machine learning, representation learning, autoencoders, CNNs, LSTMs, GANs, and advanced topics such as zero-shot learning and model robustness.

Advanced Topics in AI (Postgraduate) — 2019–2021 — IPM

Main Lecturer Trustworthiness: safety, security, fairness, transparency

Focused on trustworthy AI: safety, security, fairness, bias, transparency, and adversarial robustness. Students led paper discussions and implementation projects; assessment was based on seminars, projects, and a final exam.

Advanced Topics in AI: Deep Learning Safety & Security (Postgraduate) — 2019 — Iran University of Science and Technology

Main Lecturer

Paralleled the IPM “Advanced Topics in AI” course, with added emphasis on adversarial attacks and model robustness.