A full graduate DL syllabus, minus the tuition

What it does

This repository is the public syllabus for a two-semester graduate course on applied deep learning. It organizes PDF lecture notes and dedicated YouTube playlists into modules covering training fundamentals, computer vision, NLP, generative networks, reinforcement learning, and advanced topics like federated and self-supervised learning. The README also links to the instructor’s accompanying arXiv survey on open problems in the field.

The interesting bit

The curriculum treats the repository as a living index, spanning everything from image classification and object detection to diffusion models, graph neural networks, and computational biology. The instructor explicitly orients the material toward industry practice, warning that deep learning moves fast and demanding that students write clean Python in TensorFlow or PyTorch.

Key highlights

• Two-semester arc: Part I (Fall) focuses on computer vision; Part II (Spring) covers NLP, generative models, RL, and advanced topics
• Each module pairs PDF lecture notes with a dedicated YouTube playlist
• Includes recent topics often missing from older courses: diffusion models, multimodal learning, AutoML, and self-supervised learning
• Accompanied by the arXiv paper Open Problems in Applied Deep Learning
• Targets graduate students, though undergraduates with strong math and coding backgrounds are also welcome

Verdict

Best for self-learners who want a structured, graduate-level roadmap with free lecture notes and videos. Skip it if you are hunting for executable notebooks, assignments, or a quick API reference.