arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1104.2097cs.LG
收藏

PAC可学习性与VC维度:统计学习基本结果的脚注

PAC learnability versus VC dimension: a footnote to a basic result of statistical learning

Universidade Federal de Santa Catarina(圣卡塔琳娜联邦大学);University of Ottawa(渥太华大学)

Vladimir Pestov

摘要:

统计学习理论的一个基本结果表明,一个概念类是PAC可学习的当且仅当它是一个均匀的Glivenko-Cantelli类当且仅当该类的VC维是有限的。然而,该定理仅在类的可测性的特殊假设下有效,在这种情况下,PAC可学习性甚至变得一致。否则,有一个经典的例子,由Dudley和Durst在连续统假设下构造,并由Blumer,Eschericht,Haussler和Warrant进一步改编,VC维1的概念类既不是均匀的Glivenko-Cantelli也不是一致的PAC可学习的。我们发现,而令人惊讶的是,在一个额外的集理论的假设,这是比连续统假设(马丁公理)温和得多,PAC的可学习性是相当于有限的VC维的每一个概念类。

英文摘要:

A fundamental result of statistical learnig theory states that a concept class is PAC learnable if and only if it is a uniform Glivenko-Cantelli class if and only if the VC dimension of the class is finite. However, the theorem is only valid under special assumptions of measurability of the class, in which case the PAC learnability even becomes consistent. Otherwise, there is a classical example, constructed under the Continuum Hypothesis by Dudley and Durst and further adapted by Blumer, Ehrenfeucht, Haussler, and Warmuth, of a concept class of VC dimension one which is neither uniform Glivenko-Cantelli nor consistently PAC learnable. We show that, rather surprisingly, under an additional set-theoretic hypothesis which is much milder than the Continuum Hypothesis (Martin's Axiom), PAC learnability is equivalent to finite VC dimension for every concept class.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑