Metraipolis › archive › dc_ff98f967f757

Delay Complexity Dimension: Learning with Delayed and Censored Labels — arXiv:2610.18935

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.18935
License (as archived)
CC0-1.0
Archived
2026-10-15T14:23:07.737Z
SHA-256 of the archived text
57942b8742ccee6a7c2ce41ec75c390e5d17f79d48d652775557809040e6b66f
Stable URL of this page
https://metraipolis.com/documento/dc_ff98f967f757

How to cite

Delay Complexity Dimension: Learning with Delayed and Censored Labels — arXiv:2610.18935 — arXiv (en), https://arxiv.org/abs/2610.18935 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ff98f967f757 (Metraipolis archive dc_ff98f967f757, sha256 57942b8742ccee6a7c2ce41ec75c390e5d17f79d48d652775557809040e6b66f)
BibTeX
@misc{dcff98f967f757,
  title = {Delay Complexity Dimension: Learning with Delayed and Censored Labels — arXiv:2610.18935},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.18935},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ff98f967f757; sha256 57942b8742ccee6a7c2ce41ec75c390e5d17f79d48d652775557809040e6b66f},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.18935. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We study PAC learning of binary classifiers when labels arrive after random delays and are censored beyond an observation window $W$. How many samples does this observation model require? We propose a distribution-dependent answer, the Delay Complexity Dimension (DCD). The DCD combines the VC dimension with the exact label-information rate of the observation channel. Every learner needs $\Omega(\mathrm{DCD}/\varepsilon^2)$ samples. An oracle pseudo-label ERM matches this rate on the instance family of the lower bound, up to constants and logarithmic factors. On that family the DCD is within a constant factor of the classical censored-ERM complexity, and whether the DCD rate is achievable where the two separate remains open. The theory also yields a quantization scheme under which delay information lowers the residual label variance, an expedition policy with a $(1-1/e)$ guarantee for its version-space objective, and a phenomenon with no analogue in the no-delay setting. Under a fixed collection horizon, a moderate window can strictly beat immediate labels. Experiments on synthetic data and the Criteo conversion logs support these predictions. A budget-matched test of a companion labeler-diversity prediction is reported as a negative result.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (stat.ML) ; Information Theory (cs.IT); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.18935 — 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 2026-10-15T13:53:35.803Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).