Full search page →

DLCV 07 · Small dynamics errors compound in rollouts

AdvancedDLCV 13 · World Models: Video, Latent Dynamics and PlanningInteractive plot

Description

Compare one-step predictions, open-loop rollout and periodic observation refresh under the same actions.

Try this: Set the model coefficient to the true value 0.96. Then change it to 1.02 and increase the horizon. How much can observation refresh correct?

Lab · DLCV 07 · Small dynamics errors compound in rollouts

One-step MSE
0.00195
Open-loop MSE
1.15919
Refreshed-rollout MSE
0.02323

One-step evaluation repeatedly supplies the true state. An open-loop rollout must consume its own imperfect predictions. Periodic observations limit drift without improving the transition model itself.

Try this: Set the model coefficient to the true value 0.96. Then change it to 1.02 and increase the horizon. How much can observation refresh correct?

True scalar state: x(t+1)=0.96x(t)+u(t), x(0)=1, u(t)=a sin(0.45t). The model replaces 0.96 with a chosen coefficient. Refresh feeds the true state every k steps; k=0 disables it. Fully observed deterministic dynamics, no learned video model or replanning.

Method sources

Open in the Playground →