Metraipolis › archive › dc_b76b71ef5d20
Right Ranking, Wrong Scale: Reproducing the Robotic World Model's Dynamics Model and Testing the Uncertainty of Its Follow-up — arXiv:2611.02699
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.02699
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T18:41:16.433Z
- SHA-256 of the archived text
cc93931c77de929ddb5981312a914f319be75bba7156de08ff20a6dd04ffc150- Stable URL of this page
https://metraipolis.com/documento/dc_b76b71ef5d20
How to cite
Right Ranking, Wrong Scale: Reproducing the Robotic World Model's Dynamics Model and Testing the Uncertainty of Its Follow-up — arXiv:2611.02699 — arXiv (en), https://arxiv.org/abs/2611.02699 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b76b71ef5d20 (Metraipolis archive dc_b76b71ef5d20, sha256 cc93931c77de929ddb5981312a914f319be75bba7156de08ff20a6dd04ffc150)
BibTeX
@misc{dcb76b71ef5d20,
title = {Right Ranking, Wrong Scale: Reproducing the Robotic World Model's Dynamics Model and Testing the Uncertainty of Its Follow-up — arXiv:2611.02699},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.02699},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b76b71ef5d20; sha256 cc93931c77de929ddb5981312a914f319be75bba7156de08ff20a6dd04ffc150},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02699. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We reimplement the dynamics model of the Robotic World Model (RWM) and test the uncertainty estimate that its follow-up, RWM-U, uses to penalize reward in offline policy optimization. Our implementation matches the released model's gradients. On held-out data, training on the model's own rollouts beats one-step training at the preregistered long horizon for all ten episodes, but loses on short one-step windows. RWM-U's ensemble disagreement ranks rollout errors better than forecast step (Spearman rho = 0.90 at 100 steps), yet its spread substantially understates error: one-sigma coverage on the released checkpoint is 4.6% at 100 steps, versus 68% for a calibrated Gaussian. Rolling ensemble members on their own predictions raises coverage to 33%. The implemented squared-error loss on sampled predictions drives each member's variance toward zero. A horizon-specific rescaling restores one-sigma coverage under leave-one-episode-out evaluation, but not two-sigma coverage. We also identify two defects in the released evaluation pipeline. The study covers one robot, one gait and one dataset; it trains no policy. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Robotics (cs.RO) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.02699 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T18:12:54.780Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).