Testing Homogeneity for Normal Mixture Models: Variational Bayes Approach

被引：0
|
作者：
Kariya, Natsuki [1 ]
Watanabe, Sumio [1 ]
机构：
[1] Tokyo Inst Technol, Dept Math & Comp Sci, Tokyo 1528552, Japan
来源：
IEICE TRANSACTIONS ON FUNDAMENTALS OF ELECTRONICS COMMUNICATIONS AND COMPUTER SCIENCES | 2020年 / E103A卷 / 11期
关键词：
hypothesis test; Bayesian statistics; variational inference; singular model; mixture model; likelihood ratio; LIKELIHOOD RATIO TEST; STOCHASTIC COMPLEXITIES; INFERENCE;
D O I：
10.1587/transfun.2019EAP1172
中图分类号：
TP3 [计算技术、计算机技术];
学科分类号：
0812 ;
摘要：
The test of homogeneity for normal mixtures has been used in various fields, but its theoretical understanding is limited because the parameter set for the null hypothesis corresponds to singular points in the parameter space. In this paper, we shed a light on this issue from a new perspective, variational Bayes, and offer a theory for testing homogeneity based on it. Conventional theory has not reveal the stochastic behavior of the variational free energy, which is necessary for constructing a hypothesis test, has remained unknown. We clarify it for the first time and construct a new test base on it. Numerical experiments show the validity of our results.
引用
收藏
页码：1274 / 1282
页数：9
相关论文
共 50 条
  • [1] TESTING FOR HOMOGENEITY IN MIXTURE MODELS
    Gu, Jiaying
    Koenker, Roger
    Volgushev, Stanislav
    [J]. ECONOMETRIC THEORY, 2018, 34 (04) : 850 - 895
  • [2] Parsimonious reduction of Gaussian mixture models with a variational-Bayes approach
    Bruneau, Pierrick
    Gelgon, Marc
    Picarougne, Fabien
    [J]. PATTERN RECOGNITION, 2010, 43 (03) : 850 - 858
  • [3] Variational Bayes for Mixture Models with Censored Data
    Kohjima, Masahiro
    Matsubayashi, Tatsushi
    Toda, Hiroyuki
    [J]. MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2018, PT II, 2019, 11052 : 605 - 620
  • [4] Testing homogeneity in gamma mixture models
    Liu, X
    Pasarica, C
    Shao, YZ
    [J]. SCANDINAVIAN JOURNAL OF STATISTICS, 2003, 30 (01) : 227 - 239
  • [5] Testing homogeneity in contaminated mixture models
    Liu, Guanfu
    Fu, Yuejiao
    Liu, Wenchen
    Mu, Rongji
    [J]. CANADIAN JOURNAL OF STATISTICS-REVUE CANADIENNE DE STATISTIQUE, 2022, 50 (03): : 1034 - 1046
  • [6] Testing homogeneity in a scale mixture of normal distributions
    Xiaoqing Niu
    Pengfei Li
    Peng Zhang
    [J]. Statistical Papers, 2016, 57 : 499 - 516
  • [7] Variational Bayes inference of spatial mixture models for segmentation
    Woolrich, Mark W.
    Behrens, Timothy E.
    [J]. IEEE TRANSACTIONS ON MEDICAL IMAGING, 2006, 25 (10) : 1380 - 1391
  • [8] Testing homogeneity in a scale mixture of normal distributions
    Niu, Xiaoqing
    Li, Pengfei
    Zhang, Peng
    [J]. STATISTICAL PAPERS, 2016, 57 (02) : 499 - 516
  • [9] Testing homogeneity in a heteroscedastic contaminated normal mixture
    Niu, Xiaoqing
    Li, Pengfei
    Fu, Yuejiao
    [J]. JOURNAL OF APPLIED STATISTICS, 2019, 46 (08) : 1478 - 1491
  • [10] Parameter-based reduction of Gaussian mixture models with a variational-Bayes approach
    Bruneau, Pierrick
    Gelgon, Marc
    Picarougne, Fabien
    [J]. 19TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOLS 1-6, 2008, : 144 - 147
12345