Open Models Almanac
Methodspost-trainingrollout & RL infrastructure

single source · post-training

One-step off-policy asynchronous reinforcement learning

Overlaps rollout generation with policy updates using a one-step off-policy asynchronous setup.

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

Also called one-step off-policy asynchronous RL setup.

Filed under post-training › rollout & RL infrastructure