Open Models Almanac
Methodstraining objectiveauxiliary loss

single mention · training objective

KL regularization to a reference policy

Penalizes policy deviations from a reference policy during reinforcement learning.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See auxiliary loss for the methods filed beside it.

Also called KL regularization term.

Filed under training objective › auxiliary loss