ConceptTheory & Algos
Testable learning
A model of learning where an algorithm runs a test on a training set, outputs a classifier with guaranteed error relative to the optimal plus an epsilon along with a certificate, and is allowed to reject when the marginal distribution is atypical.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Lecturers
1Connections
5The lectures that define it or assume it, and the concepts next to it.