Metraipolis › archive › dc_d2e2ceff5de0

Validation-Frontier Representation Selection under Constrained Observation — arXiv:2608.15095

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15095
License (as archived)
CC0-1.0
Archived
2026-10-18T17:12:55.158Z
SHA-256 of the archived text
668d15c19a2a67304631bf44860f8b380f111adad9be49f30f7b805e6732bde9
Stable URL of this page
https://metraipolis.com/documento/dc_d2e2ceff5de0

How to cite

Validation-Frontier Representation Selection under Constrained Observation — arXiv:2608.15095 — arXiv (en), https://arxiv.org/abs/2608.15095 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2e2ceff5de0 (Metraipolis archive dc_d2e2ceff5de0, sha256 668d15c19a2a67304631bf44860f8b380f111adad9be49f30f7b805e6732bde9)
BibTeX
@misc{dcd2e2ceff5de0,
  title = {Validation-Frontier Representation Selection under Constrained Observation — arXiv:2608.15095},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15095},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2e2ceff5de0; sha256 668d15c19a2a67304631bf44860f8b380f111adad9be49f30f7b805e6732bde9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15095. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

AI systems deployed outside clean benchmark settings often rely on observations that are incomplete, unstable, costly, or degraded by monitoring failures. This paper studies representation selection under constrained observation: choosing a state representation when raw accuracy is not the only operational criterion. We propose a validation-frontier selector that combines balanced accuracy with penalties for feature cost, overfit gap, and validation-test instability. In a focused public-tabular benchmark using three scikit-learn datasets, five observation regimes, 45 matched task cells, 720 candidate actions, and 405 representation rows, the adaptive selector improves frontier score over full trace features by 0.025801 while reducing mean feature count by 22.733. Balanced-accuracy difference is small and not statistically significant. A broader offline stress test gives mixed results. The supported claim is therefore bounded: adaptive representation selection can improve a constrained-observation robustness-efficiency frontier in matched benchmark settings, but does not universally dominate trace baselines.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.15095 — 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 2026-10-18T16:29:13.767Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).