Metraipolis › archive › dc_1b2dfda7963f
An Optimal Agnostic PAC Algorithm — arXiv:2608.06363
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.06363
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:50:23.632Z
- SHA-256 of the archived text
595b06cd1be8f331641787186a89c97e08446f3c1d50afc43c65d5df258c1ab3- Stable URL of this page
https://metraipolis.com/documento/dc_1b2dfda7963f
How to cite
An Optimal Agnostic PAC Algorithm — arXiv:2608.06363 — arXiv (en), https://arxiv.org/abs/2608.06363 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_1b2dfda7963f (Metraipolis archive dc_1b2dfda7963f, sha256 595b06cd1be8f331641787186a89c97e08446f3c1d50afc43c65d5df258c1ab3)
BibTeX
@misc{dc1b2dfda7963f,
title = {An Optimal Agnostic PAC Algorithm — arXiv:2608.06363},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.06363},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_1b2dfda7963f; sha256 595b06cd1be8f331641787186a89c97e08446f3c1d50afc43c65d5df258c1ab3},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.06363. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Let $H\subseteq\{-1,+1\}^X$ be a class of finite VC dimension $d\ge1$. Writing $L$ for the binary risk and $L^*=\min_{h\in H}L(h)$, we construct a learner achieving the statistically optimal risk bound: from an i.i.d.\ sample of size $n$, for every $0<\delta\le 1/2$, with probability at least $1-\delta$, \[ L(\widehat h) \le L^*+ 7\cdot10^8\left( \sqrt{\frac{L^*(d+\log(1/\delta))}{n}} +\frac{d+\log(1/\delta)}{n} \right). \] This settles the sample complexity of agnostic PAC learning up to universal constants at every fixed $L^*$, matching the lower bounds of Devroye, Györfi, and Lugosi [A Probabilistic Theory of Pattern Recognition, Springer, 1996].
Tema degli esploratori: matematica e informatica teorica (retro-catalogo arXiv: annunci del 2026-08-07 in cs.DS).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Data Structures and Algorithms (cs.DS); Statistics Theory (math.ST).
Pagina del lavoro: https://arxiv.org/abs/2608.06363 — 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 2026-10-31T10:37:54.281Z dalla pagina https://arxiv.org/catchup/cs.DS/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).