不可知直接和的速率分离
A Rate Separation for Agnostic Direct Sums
- Truth Audit Labs
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
针对不可知PAC学习，研究直接和$C^r$的学习曲线与单实例学习曲线的依赖关系，证明单实例学习速率不决定直接和速率，并给出两类函数的不可知学习曲线阶数结果。
中文摘要 AI 辅助
Hanneke、Moran和Waknine[1]提出，不可知PAC学习中，直接和$C^r$的学习曲线如何依赖于单实例学习曲线$\boldsymbol{\text{epsagn}}(n\boldsymbol{\boldsymbol{\text{|}}} C)$和$r$。我们证明单实例学习速率不决定直接和速率。设$\boldsymbol{\text{F}}$为两个常值二元函数类，$\boldsymbol{\text{G}}$为零函数与恒等函数构成的类，两类的不可知学习曲线均为$n^{-1/2}$阶。
英文摘要
Hanneke, Moran, and Waknine \cite{HannekeMoranWaknine2024} asked how the agnostic PAC learning curve of the direct sum $C^r$ depends on the single-instance learning curve $\epsagn(n\mid C)$ and on $r$. We show that the single-instance learning rate does not determine the direct-sum rate. Let $\F$ be the class of the two constant binary functions and let $\G$ consist of the zero function and the identity function. Both classes have agnostic learning curve of order $n^{-1/2}$.
