ConceptAI & ML
Evidence lower bound
A lower bound to the data likelihood under a model, used as an objective in variational autoencoders.
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
2Each in their own terms, from the easiest level.
Sergey Levine
A lower bound on the log probability of data that is maximized to perform variational approximation.
Connections
7The lectures that define it or assume it, and the concepts next to it.