ConceptAI & ML
Batch reinforcement learning via importance sampling
A technique for offline reinforcement learning that uses importance sampling to adjust policy gradients calculated from a behavior policy to evaluate or optimize a target policy.
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.