AWhat changed
Learned policies and values by propagating gradients through imagined trajectories in a compact learned dynamics model, with visual-control evaluations.
Scientific scope
Learned policies and values by propagating gradients through imagined trajectories in a compact learned dynamics model, with visual-control evaluations. The policy is optimized in imagination but the world model is trained on real environment experience; this is not data-free control. First preprint 2019, cited ICLR version 2020. Model error and finite benchmarks limit generalization.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. The policy is optimized in imagination but the world model is trained on real environment experience; this is not data-free control. First preprint 2019, cited ICLR version 2020. Model error and finite benchmarks limit generalization.
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.
- Dream to Control: Learning Behaviors by Latent Imagination
- Hafner et al.: Dreamer latent-imagination behavior learning — author/company project record
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