PETS: probabilistic dynamics ensembles with trajectory sampling

What changed

Combined bootstrapped probabilistic dynamics models with particle trajectory sampling and model-predictive control, distinguishing transition noise from model disagreement during planning.

Scientific scope

Combined bootstrapped probabilistic dynamics models with particle trajectory sampling and model-predictive control, distinguishing transition noise from model disagreement during planning. Ensemble and Gaussian uncertainty are modeling proxies, not calibrated error certificates. TS1 and TS-infinity propagate model uncertainty differently. Reported environment-sample efficiency does not imply lower total compute or guarantees outside the tested settings.

Evidence

The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Ensemble and Gaussian uncertainty are modeling proxies, not calibrated error certificates. TS1 and TS-infinity propagate model uncertainty differently. Reported environment-sample efficiency does not imply lower total compute or guarantees outside the tested settings.

Contributors

• Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.

• Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Models

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction