How to revise Introduction to Deep Learning and Generative AI
Introduction to Deep Learning and Generative AI questions reward clear reasoning as well as a final answer. This revision guide is a starting point for using the available papers; it is not the official syllabus or a promise about future exams.
Revision focus
Neural networks; backpropagation; optimisation; representation learning
Useful prior knowledge
Matrix algebra, derivatives and probability
Common mistakes
Losing a tensor dimension or applying a gradient without the chain rule
Use these papers effectively
Write each tensor shape and the loss before differentiating. Verify one gradient path and distinguish training behaviour from inference. After checking a solution, solve the question again without looking and record the specific step you missed.
Revision guidance only. Check your current official course materials for the assessed syllabus.