Bayesian Distance Weighted Discrimination.

Bayesian Distance Weighted Discrimination.
复制标题

DOI:
10.1080/10618600.2022.2069778
复制
发表时间:
2022
期刊:
Journal of computational and graphical statistics : a joint publication of American Statistical Association, Institute of Mathematical Statistics, Interface Foundation of North America
影响因子:
--
通讯作者:
--
中科院分区:
其他
文献类型:
--
作者:

文献摘要

参考文献

相似文献

距离加权判别(DWD)是一种特别适合于高维数据分类任务的线性判别方法。DWD系数最小化了一个直观的目标函数,可以使用最先进的优化技术有效地求解。然而,DWD尚未被纳入基于模型的统计推断框架。在本文中,我们展示了DWD识别适当的贝叶斯后验分布的模式,这是由类概率的特定链接函数和系数的收缩性适当先验分布产生的。我们描述了一个相对有效的马尔可夫链蒙特卡罗(MCMC)算法,在这个贝叶斯框架下从真后验进行模拟。我们证明了后验是渐近正态的,并推导了其极限分布的均值和协方差矩阵。通过几项模拟研究和对乳腺癌基因组学的应用,我们展示了贝叶斯方法如何用于(1)计算校准良好的后验类概率,(2)评估DWD系数和所得样本分数的不确定性,(3)在并非所有类标签可用时通过半监督分析提高功率,以及(4)在基于模型的框架内自动确定惩罚调优参数。执行贝叶斯DWD的R代码可在https://github.com/lockEF/BayesianDWD获得。
Distance weighted discrimination (DWD) is a linear discrimination method that is particularly well-suited for classification tasks with high-dimensional data. The DWD coefficients minimize an intuitive objective function, which can solved efficiently using state-of-the-art optimization techniques. However, DWD has not yet been cast into a model-based framework for statistical inference. In this article we show that DWD identifies the mode of a proper Bayesian posterior distribution, that results from a particular link function for the class probabilities and a shrinkage-inducing proper prior distribution on the coefficients. We describe a relatively efficient Markov chain Monte Carlo (MCMC) algorithm to simulate from the true posterior under this Bayesian framework. We show that the posterior is asymptotically normal and derive the mean and covariance matrix of its limiting distribution. Through several simulation studies and an application to breast cancer genomics we demonstrate how the Bayesian approach to DWD can be used to (1) compute well-calibrated posterior class probabilities, (2) assess uncertainty in the DWD coefficients and resulting sample scores, (3) improve power via semi-supervised analysis when not all class labels are available, and (4) automatically determine a penalty tuning parameter within the model-based framework. R code to perform Bayesian DWD is available at https://github.com/lockEF/BayesianDWD.
DOI: 10.1080/10618600.2012.700878
发表时间: 2013-12-01
影响因子: 2.4
作者:
Huang, Hanwen;Liu, Yufeng;Marron, J. S.
通讯作者: Marron, J. S.
DOI: 10.1023/a:1012489924661
发表时间: 2002-01-01
期刊: MACHINE LEARNING
影响因子: 7.5
作者:
Sollich, P
通讯作者: Sollich, P
DOI: 10.1080/10618600.2015.1049700
发表时间: 2016-09-01
影响因子: 2.4
作者:
Wang, Boxiang;Zou, Hui
通讯作者: Zou, Hui
DOI: 10.1080/00401706.2018.1529629
发表时间: 2019-07-03
期刊: TECHNOMETRICS
影响因子: 2.5
作者:
Wang, Boxiang;Zou, Hui
通讯作者: Zou, Hui
DOI: 10.1214/11-ba601
发表时间: 2011-01-01
期刊: BAYESIAN ANALYSIS
影响因子: 4.4
作者:
Polson, Nicholas G.;Scott, Steven L.
通讯作者: Scott, Steven L.