English(EN) Near-Optimal List Replicability from Littlestone Dimension

新理论收紧了PAC学习中列表可复现性的界限

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-27 04:00

研究人员在PAC学习框架中为列表可复现性建立了一个新的上限，显著改进了先前的结果。研究表明，对于具有Littlestone维度为'd'和对偶VC维度为'v*'的概念类，列表可复现性由$2^{O(d \min\{\log d, v^*\})}$界定。这个新界限接近最优，紧密匹配已知的下界$2^{\Omega(d)}$，并将先前最佳上限$2^{2^{O(d)}}$指数级地改进。这些发现源于一种新颖的归纳框架，并利用了差分隐私学习和稳定正则化的工具。 AI

影响建立了与机器学习算法设计和分析相关的新理论界限。

排序理由这是一篇发表在arXiv上的理论计算机科学论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 8 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 这是一篇发表在arXiv上的理论计算机科学论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Same-day; Cluster formed today. Ranking reflects the current source set at time of score.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Ari Blondal, Pooya Hatami, Chavdar Lalov, Sivan Tretiak · 2026-10-27 04:00

Littlestone维度近乎最优的列表可复现性

arXiv:2610.32853v1 Announce Type: new Abstract: We study the list replicability number in the realizable PAC learning framework and prove that every concept class $H$ over a countable domain with Littlestone dimension $d$ and dual VC dimension $v^*$ satisfies \[ \mathrm{LR}(H)\le…

本文如何被排名

报道来源 [1]

Littlestone维度近乎最优的列表可复现性

相关实体

相关话题