specific method · filed under post-training
One-step off-policy asynchronous reinforcement learning
Overlaps rollout generation with policy updates in a one-step off-policy asynchronous training setup.
Also called one-step off-policy asynchronous RL setup.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1core 1
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
During RL and MOPD, we train using a one-step off-policy asynchronous RL setup, so rollout generation is overlapped with the policy update
coreoptimizationin Nemotron 3 UltraNVIDIA
we train using a one-step off-policy asynchronous RL setup, so rollout generation is overlapped with the policy update
usedunclearin Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under post-training :: rollout & RL infrastructure.
Asynchronous reinforcement learningPartial rolloutAsynchronous RL frameworks for large-scale agent scaffolds and environment orchestrationToken-in-token-out (TITO)Asynchronous reinforcement learning infrastructureCo-located RL trainingData SchedulerDecoupled control plane and data planeDecoupling agent rollout into sandbox and worker containerDeficit-corrected schedulingLarge-scale asynchronous RL in synthesized tasksPredictive Rollout DispatchSample-grained garbage collectionSeamless Rollout EngineSLIMEToken-granularity persistence of rollout statesToken-level interruptionTool ManagerToolboxAdaptive Rollout ConcurrencyAdaptive Rollout SchedulingAgent LoopAgent-centric rollout executionAsynchronous Agent RL algorithms