Unverified50% confidenceFactExact time
On-policy rollout means sampling generated sequences under the distribution of the current policy (i.e., the student model itself), rather than using offline data produced by a fixed dataset or an old policy.
1
Sources
50%
Confidence
Long-term
Relevance
10/26/2026
First Seen
Sources
Related Entities
Related Claims
Unverified策略梯度方法本质上是on-policy的，因为梯度估计中的期望是关于当前策略分布的69% similarUnverifiedIn the reinforcement learning pipeline, the model continuously generates candidate outputs (rollouts), which are evaluated and scored before being fed back into the training process to update the policy.67% similarUnverifiedMeans of mitigating distribution shift include rolling backtests on recent data, introducing exogenous variables, and online calibration of model outputs; the fundamental measure is to establish monitoring that triggers human review or model updates when the distribution of forecast residuals changes significantly.66% similarUnverifiedUsing generative model rollouts as a data source carries a risk of distribution shift: the generative model's own physical priors may contain biases or simplifications, making the data unable to fully reflect the complex dynamics of the real world.64% similarUnverifiedRolling preview refers to a model's continuous iteration by frequently pushing checkpoints before official release, where capabilities between adjacent versions may exhibit inconsistent fluctuations rather than monotonic improvement.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1098819API
curl https://kongchang.com/api/v1/knowledge/claims/1098819MCP
get_claim(id=1098819)