• deep learning
• cyber security
• seminars

Note: I like to keeps the slides fairly minimal and talk a lot during the lectures. If you have missed a lecture, please listen to the Encore recordings.

Deep Learning

• Lecture 1: Intelligence and learning - from nature to machine. (slides)
• Lecture 2: PyTorch Programming - live programming tutorial with minimal slides. (slides)
• Lecture 3: Building and improving architectures. (slides)
• Lecture 4: Neural Networks and Backpropagation: by Grégoire Payen de La Garanderie. (slides, handout)
• Lecture 5: Density Estimation and Generative Models. (slides)
• Lecture 6: Advances in Generative Models. (slides)
• Lecture 7: Reinforcement Learning. (slides)
• Lecture 8: Recurrent Neural Networks: by Phil Jackson. (slides)
• Lecture 9: Clustering and Manifold Learning. (slides)
• Lecture 10: Learning to Learn within the bigger picture. (slides)

Deep Learning Teaching Materials

• PyTorch lecture materials
• Simple network on FashionMNIST
• CNN on CIFA100
• Simple GAN
• Wasserstein GAN
• Variational Autoencoder
• UNet

Cyber Security

• Lecture 1: Introduction, history, today, and terminology slides. (slides)
• Lecture 2: Applied cryptography slides. (slides)
• Lecture 3: Operating system security and access control slides. (slides)
• Lecture 4: Identification, authentication, and authorization: by Grégoire Payen de La Garanderie*. (slides)
• Lecture 5: Network and web security slides. (slides)
• Lecture 6: Database security slides. (slides)
• Lecture 7: Software security - understanding the platform slides. (slildes)
• Lecture 9: Security in industry: by Jonathan Frawley (slides)
• Lecture 10: Software security and AI - exploits, mitigations, and the future. (slides)

Please contact me for permission in using the lab material in other courses:

• Practical 1: Hacking and securing a simple web server. (TBA)

• Practical 1: Hacking and securing a simple web server. (pdf, materials)
• Practical 2: Traversal and XSS attacks on a web server. (pdf, materials)
• Practical 3: Database security. (pdf, materials)
• Practical 4: Software security (this lab involves a 4.7gb virtual machine prepared by Ross Bradley at SRM).
• Answers 1: Sample answers for some of the tasks in practical 1. (pdf)
• Answers 2: Sample answers for some of the tasks in practical 2. (pdf)
• Answers 3: Sample answers for some of the tasks in practical 3. (pdf)

Research Seminars

• Machine learning with problematic datasets in diverse applications (CUHK)
• How to write a good literature review quickly. (PGR seminar)
• Active implicits: segmentation of complex & noisy images. (DBIL seminar)
• GPU-based technology for fast segmentation in 3D imaging data. (DBIL seminar)
• Automated discovery: classification from semi-structured data sources. (Industry seminar)
• Deep Learning: guide for classification, regression, and clustering of big datasets. (Durhack)
• High-level GPGPU programming with OpenCL and CUDA, applied to real-time photorealistic rendering and deformation. (invited seminar, included live-coding session)
• Describing & segmenting images with implicit shape functions - research day (won 2^nd best talk).
• Feature-varying skeletonization (CGI).
• Hierarchical unbounded signed-distance fields. (Newcastle University)
• Sparse volumetric deformation: real-time deformation and rendering of massive amounts of volumetric data on current hardware. (PGR seminar)

* seminar slides are available on request.