Postgraduate / Master's · Artificial Intelligence
Machine Learning
What is it
Postgraduate machine learning moves beyond applying standard algorithms to understanding their theoretical foundations — statistical learning theory, generalisation bounds, optimisation landscapes, and the mathematical assumptions underlying why and when different learning methods succeed or fail.
Why it matters
At research level, simply knowing how to call a library function is not enough — advancing the field, diagnosing subtle model failures, or adapting methods to genuinely novel problems all require understanding the theory well enough to reason about a method's behaviour outside the textbook cases it was originally designed for.
Exam tip
When reviewing a research paper's proposed method, look specifically for what assumptions it makes about the data or problem structure — a method's headline results often hold only under conditions that are easy to overlook if you focus purely on the reported accuracy figures.
Related topics
Want help mastering Machine Learning?
Tell us about the student's goals and confidence — we'll design a personalised plan.
