Statistical Learning Theory
Why fitting a sample tells you anything about the world it was drawn from, what the capacity of a model class really measures, and the theorem that says no method is best everywhere - together with what that theorem does not say.
Sign in to take quizzes, earn XP, and unlock stages as you reach 90% mastery.
Generalisation and the Union Bound
30 min · 120 XPThe gap between the error you measure and the error you will suffer, why choosing the best of many candidates makes that gap grow, and the counting argument that turns it into a guarantee.
Open lesson →Sign in to take the 3-question quiz.
VC Dimension and What It Measures
30 min · 120 XPCapacity for infinite classes, defined by the labellings a family can realise rather than by counting its members, found by searching for the largest shattered set, and the lemma that makes it useful.
Open lesson →Sign in to take the 3-question quiz.
No Free Lunch and What It Does Not Say
25 min · 100 XPThe averaging argument that equalises every learner over all possible targets, verified by enumerating them, and the reason it is a statement about assumptions rather than a counsel of despair.
Open lesson →Sign in to take the 3-question quiz.