ConceptAI & ML
Q-learning overestimation
A systematic bias where Q-values estimated by Q-learning are consistently higher than the actual discounted returns due to the max operation over noisy value estimates.
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
4The lectures that define it or assume it, and the concepts next to it.