ConceptAI & ML
Residual gradient algorithms
A class of methods that properly apply the chain rule to the Q-learning objective, though they tend to work poorly in practice due to numerical ill-conditioning.
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.