ConceptAI & ML
Truncated backpropagation through time
An approximation method where the sequence is broken into chunks, carrying hidden states forward indefinitely while only backpropagating through a limited number of steps.
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.