ConceptAI & ML
Importance-weighted policy gradient estimator
An estimator for policy gradients in language models that uses importance sampling to reuse samples generated from a previous 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.