Asymptotic comparison of semi-supervised and supervised linear discriminant functions for heteroscedastic normal populations

Asymptotic comparison of semi-supervised and supervised linear discriminant functions for heteroscedastic normal populations
复制标题

异方差正态总体的半监督和监督线性判别函数的渐近比较

DOI:
10.1007/s11634-016-0266-6
复制
发表时间:
2018
影响因子:
1.6
通讯作者:
Kenichi Hayashi
Kenichi Hayashi
中科院分区:
计算机科学3区
文献类型:
--
作者:
Yuki Kawakubo;Tatsuya Kubokawa and Muni S. Srivastava;足立浩平;Norihiro Kamide;Kenichi Hayashi

文献摘要

参考文献

相似文献

已有研究表明,将未标记数据和标记数据结合起来构造判别函数在实际应用中取得了成功。然而,理论研究表明,未标记的数据有时会对判别函数的性能产生不利影响。因此,了解哪些情况需要使用未标记数据非常重要。在异方差正态性假设下,本文提出了渐近相对效率作为比较有无未标记数据分析的度量。考虑了使受试者工作特征曲线下面积最大化的线性判别函数。渐近相对效率的评估,调查何时以及如何未标记的数据有助于提高判别性能在几种条件下。结果表明,渐近相对有效性主要取决于协方差矩阵的异方差性和观察样本标签的随机结构。
It has been reported that using unlabeled data together with labeled data to construct a discriminant function works successfully in practice. However, theoretical studies have implied that unlabeled data can sometimes adversely affect the performance of discriminant functions. Therefore, it is important to know what situations call for the use of unlabeled data. In this paper, asymptotic relative efficiency is presented as the measure for comparing analyses with and without unlabeled data under the heteroscedastic normality assumption. The linear discriminant function maximizing the area under the receiver operating characteristic curve is considered. Asymptotic relative efficiency is evaluated to investigate when and how unlabeled data contribute to improving discriminant performance under several conditions. The results show that asymptotic relative efficiency depends mainly on the heteroscedasticity of the covariance matrices and the stochastic structure of observing the labels of the cases.
DOI: 10.1198/jasa.2009.tm08273
发表时间: 2009-12-01
影响因子: 3.7
作者:
Boldea, Otilia;Magnus, Jan R.
通讯作者: Magnus, Jan R.
DOI: 10.1007/s10463-009-0264-y
发表时间: 2011-10-01
影响因子: 1
作者:
Komori, Osamu
通讯作者: Komori, Osamu
DOI: 10.1006/jtbi.1995.0242
发表时间: 1995
影响因子: 2
作者:
J. Airoldi;Bernard D. Flury;M. Salvioni
通讯作者: M. Salvioni
DOI: 10.1080/01621459.1978.10480106
发表时间: 1978
影响因子: 3.7
作者:
Terence J. O'Neill
通讯作者: Terence J. O'Neill
DOI: 10.1080/00949659508811689
发表时间: 1995
影响因子: 1.2
作者:
G. McLachlan;D. Scot
通讯作者: D. Scot