Metraipolis › archive › dc_3ce6a6b97ddd
Probably Safe Sim-to-Real Transfer with Lifetime Validity — arXiv:2611.09344
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09344
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:41:33.440Z
- SHA-256 of the archived text
05e33eb5cbe9e48ab2b6a4b32006883cfaf249c036b66a5a892b98d11e9da699- Stable URL of this page
https://metraipolis.com/documento/dc_3ce6a6b97ddd
How to cite
Probably Safe Sim-to-Real Transfer with Lifetime Validity — arXiv:2611.09344 — arXiv (en), https://arxiv.org/abs/2611.09344 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3ce6a6b97ddd (Metraipolis archive dc_3ce6a6b97ddd, sha256 05e33eb5cbe9e48ab2b6a4b32006883cfaf249c036b66a5a892b98d11e9da699)
BibTeX
@misc{dc3ce6a6b97ddd,
title = {Probably Safe Sim-to-Real Transfer with Lifetime Validity — arXiv:2611.09344},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09344},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3ce6a6b97ddd; sha256 05e33eb5cbe9e48ab2b6a4b32006883cfaf249c036b66a5a892b98d11e9da699},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09344. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We present Dejavu, a method that makes sim-to-real transfer of reinforcement learning policies in constrained Markov decision processes probably safe over an unbounded sequence of policy updates. An anytime valid test using per-decision importance sampling and a betting confidence sequence certifies each candidate trained in simulation before deployment using only trajectories collected by a safe baseline policy operating in production. Simulation quality does not affect the safety guarantee, only the production sample efficiency of the method. Finally, a summable per-test failure probability yields the lifetime valid property. Empirical results across three sim-to-real benchmarks support the theoretical work. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2611.09344 — 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 2026-11-09T22:33:47.157Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).