Metraipolis › archive › dc_7ab15accf757

Sequential Recovery Updates for Learned Controllers under Successful-Trajectory Preservation — arXiv:2611.02555

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.02555
License (as archived)
CC0-1.0
Archived
2026-11-03T10:58:29.096Z
SHA-256 of the archived text
3aa6b810afe717d3989c284cb8fa2fd1b4a9076caa8db22da2cd4e33f023f407
Stable URL of this page
https://metraipolis.com/documento/dc_7ab15accf757

How to cite

Sequential Recovery Updates for Learned Controllers under Successful-Trajectory Preservation — arXiv:2611.02555 — arXiv (en), https://arxiv.org/abs/2611.02555 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7ab15accf757 (Metraipolis archive dc_7ab15accf757, sha256 3aa6b810afe717d3989c284cb8fa2fd1b4a9076caa8db22da2cd4e33f023f407)
BibTeX
@misc{dc7ab15accf757,
  title = {Sequential Recovery Updates for Learned Controllers under Successful-Trajectory Preservation — arXiv:2611.02555},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.02555},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7ab15accf757; sha256 3aa6b810afe717d3989c284cb8fa2fd1b4a9076caa8db22da2cd4e33f023f407},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.02555. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A competent learned controller already recovers from many disturbances. Improving it means learning a correction, and learning when not to apply it. Freezing a skill's parameters does not settle the second half. In a controller that routes among frozen skills, a new entry rule can redirect the states an old skill succeeded on. We therefore constrain entry instead of parameters. Each update searches for a feedback skill at a measured failure. It fits an entry rule that reads observations and rejects every state on an accumulated successful trajectory. The skill is admitted only if the composed controller reproduces the new recovery on its own and replays every protected trajectory exactly. A finite-trajectory condition states when this preserves the previous controller. One mechanism runs on four streams: two motions on each of two humanoid systems. Two updates per stream add 62 recoveries over 192 withheld perturbations, and lose none of the base controller's successes. We then price that guarantee. Removing the constraint recovers 33 more conditions but stops eight protected ones from completing at all, and excluding only what admission proves necessary recovers none of that cost.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Robotics (cs.RO) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2611.02555 — 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 2026-11-03T10:49:38.656Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).