ConceptAI & ML
Inverted Dropout
A variation of dropout where the scaling by the probability p is done during training time by dividing by p, leaving test time unchanged.
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
2The lectures that define it or assume it, and the concepts next to it.