arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 0904.4527cs.LG
收藏

先验近似忽略下范畴潜变量的学习极限

Limits of Learning about a Categorical Latent Variable under Prior Near-Ignorance

Istituto Dalle Molle di Studi sull’Intelligenza Artificiale (IDSIA)(达莱莫莱人工智能研究所);University of St. Gallen(圣加仑大学);Australian National University(澳大利亚国立大学);NICTA(澳大利亚国家信息与通信技术研究院)

Alberto Piatti, Marco Zaffalon, Fabio Trojani, Marcus Hutter

摘要:

在本文中,我们认为连贯的理论(认知)的不确定性的瓦利,其中的信念表示通过一组概率分布,我们专注于建模先验无知的问题关于一个分类随机变量。在这种情况下,一个已知的结果是,先前的无知状态与学习是不相容的。为了克服这个问题,另一种状态的信念,称为\emph{near-ignorance},已被提出。近乎无知与无知非常相似,通过满足一些可以被认为是无知状态下必要的原则,并允许学习发生。本文所做的,是提供新的和实质性的证据,也接近无知不能真正被视为一种出路的问题,开始统计推断的条件下,非常弱的信念。该结果的关键是关注以感兴趣的变量\emph{latent}为特征的设置。我们认为,这样的设置是迄今为止最常见的情况下,在实践中,我们提供的情况下,分类潜变量(和一般\emph{manifest}变量)的条件,如果满足,防止学习发生在以前的近乎无知。即使在最常见的统计问题中,这个条件也很容易满足。我们认为这些结果是一个强有力的证据,反对在实际的统计问题中采用事先近乎无知的条件。

英文摘要:

In this paper, we consider the coherent theory of (epistemic) uncertainty of Walley, in which beliefs are represented through sets of probability distributions, and we focus on the problem of modeling prior ignorance about a categorical random variable. In this setting, it is a known result that a state of prior ignorance is not compatible with learning. To overcome this problem, another state of beliefs, called \emph{near-ignorance}, has been proposed. Near-ignorance resembles ignorance very closely, by satisfying some principles that can arguably be regarded as necessary in a state of ignorance, and allows learning to take place. What this paper does, is to provide new and substantial evidence that also near-ignorance cannot be really regarded as a way out of the problem of starting statistical inference in conditions of very weak beliefs. The key to this result is focusing on a setting characterized by a variable of interest that is \emph{latent}. We argue that such a setting is by far the most common case in practice, and we provide, for the case of categorical latent variables (and general \emph{manifest} variables) a condition that, if satisfied, prevents learning to take place under prior near-ignorance. This condition is shown to be easily satisfied even in the most common statistical problems. We regard these results as a strong form of evidence against the possibility to adopt a condition of prior near-ignorance in real statistical problems.

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。

↑