ConceptMathematics
Universality theorem for neural networks
The principle that any continuous function can be approximated arbitrarily well using compositions of linear layers and ReLUs with enough folds.
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.
Lecturers
1Connections
3The lectures that define it or assume it, and the concepts next to it.