Metraipolis › archive › dc_caaa1c8b660b

Active Feature Acquisition With Incomplete Training Data — arXiv:2609.32325

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.32325
License (as archived)
CC0-1.0
Archived
2026-10-07T21:00:04.733Z
SHA-256 of the archived text
609aec00c6a7b5f268960a973fc38c3f3f70fae5bb8c5817abb9393a85a0f364
Stable URL of this page
https://metraipolis.com/documento/dc_caaa1c8b660b

How to cite

Active Feature Acquisition With Incomplete Training Data — arXiv:2609.32325 — arXiv (en), https://arxiv.org/abs/2609.32325 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_caaa1c8b660b (Metraipolis archive dc_caaa1c8b660b, sha256 609aec00c6a7b5f268960a973fc38c3f3f70fae5bb8c5817abb9393a85a0f364)
BibTeX
@misc{dccaaa1c8b660b,
  title = {Active Feature Acquisition With Incomplete Training Data — arXiv:2609.32325},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.32325},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_caaa1c8b660b; sha256 609aec00c6a7b5f268960a973fc38c3f3f70fae5bb8c5817abb9393a85a0f364},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.32325. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In many prediction tasks, acquiring all features can be a prohibitively expensive or outright impossible task. Further, in many cases a static subset of features may not be enough to solve the problem sufficiently across various instances. Active Feature Acquisition (AFA) addresses these problems by formalizing the trade-off between feature cost and predictive performance during sequential feature selection. However, prior AFA work largely assumes access to complete training data, an assumption that is often violated in practice. Here we study AFA with Incomplete Training Data (AFA-ITD), showing that under missing completely at random (MCAR) data, one-step acquisition values remain unchanged, whereas multi-step values can decrease. We analyze three approaches to learning from incomplete data: aliasing, filtering, and generative restoration. We show that filtering can require a number of training instances scaling exponentially with the dimension, whereas generative restoration scales exponentially with the acquisition budget. We empirically test our theory on a controlled experiment and across common AFA datasets and find that missingness mainly damages methods that exploit multi-step acquisitions and that generative restoration is able to recover lost performance in many experiments. Code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.32325 — 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 2026-10-07T20:29:02.294Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).