Variational Bayesian Learning for Beta Mixture Model and Its Application

LAI Yu-ping; GAO Ning; HE Wen-da; PING Yuan; Du Chun-lai; WANG Bao-cheng; DING Hong-wei

doi:10.3969/j.issn.0372-2112.2018.07.036

您当前的位置：

首页 >

文章列表页 >

Variational Bayesian Learning for Beta Mixture Model and Its Application

更新时间：2025-07-16

- Variational Bayesian Learning for Beta Mixture Model and Its Application
- Acta Electronica Sinica Vol. 46, Issue 7, Pages: 1787-1792(2018)
- 作者机构：
  
  1. 北方工业大学计算机学院,北京,100144
  2. 河南许昌学院,河南,许昌,461000
  3. 云南大学信息学院,云南,昆明,650091
  4. 北方工业大学计算机学院,北京,100144
  5. 河南许昌学院,河南,许昌,461000
  6. 云南大学信息学院,云南,昆明,650091
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China (No.61461053, No.61461053, No.61364012);Excellent Talents Support Program of North China University of Technology;Science and Technology Innovation Talent Program of Colleges and Universities of Henan Province (18HASTIT022);Science and Technology Innovative Talents Project in University of Henan Province (184100510012);Technology Research and Development Program Fund of Henan Province (182102210123);Key Project of Science and Technology Research of Education Department of Henan Province (16A520025, 18A520047)
- DOI：10.3969/j.issn.0372-2112.2018.07.036
  CLC： TP391.4
- Published Online：25 July 2018，
  
  Published：2018
- 稿件说明：
移动端阅览
LAI Yu-ping, GAO Ning, HE Wen-da, et al. Variational Bayesian Learning for Beta Mixture Model and Its Application[J]. Acta Electronica Sinica, 2018, 46(7): 1787-1792.
DOI：

LAI Yu-ping, GAO Ning, HE Wen-da, et al. Variational Bayesian Learning for Beta Mixture Model and Its Application[J]. Acta Electronica Sinica, 2018, 46(7): 1787-1792. DOI： 10.3969/j.issn.0372-2112.2018.07.036.

摘要

贝塔混合模型（Beta Mixture Model，BMM）是一种重要的非高斯概率模型，常用于有界数据的统计分析.但是由于其表达式复杂，BMM的参数估计比较困难.针对该问题，本文提出一种高效的变分贝叶斯学习方法进行参数估计.该方法采用形式简单的自由分布，通过不断最大化初始变分目标函数的下界，迭代逼近得到真实的贝叶斯后验分布.在合成数据集与实际数据集上进行实验，实验结果证明了所提出算法的有效性和可行性.

Abstract

Beta mixture model (BMM) is an important non-Gaussian probability model

which has been widely used in statistical analysis of the bounded data.It is hard to perform parameter estimation for BMM

due to its complex function format.An efficient variational Bayesian learning method has been proposed to deal with this problem.With the variational distribution and by iteratively maximizing the lower bound of the original variational object function

the approximating distribution which is the closest to the true Bayesian posterior distribution is obtained.Both synthetic and real data are experimented to demonstrate the effectiveness and the merits of the proposed approach.

关键词

Keywords

references

Views

309

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Variational Learning for Finite Inverted Dirichlet Mixture Models and Applications

A New Model Selection Criterion Based on Kullback-Leibler Information Divergence and Geometric Curvature

Multiresolution Image Denoising via Bayesian Estimator

Variational Learning for Finite Beta-Liouville Mixture Models and Its Application

Compressed Domain Speech Enhancement Based on Gaussian Mixture Model

Related Author

YANG Yi-xian

GUO Chun

GUO Yu-cui

DING Hong-wei

ZHOU Ya-jian

LAI Yu-ping

LIU Yun-hui

LUO Si-wei

Related Institution

School of Science Beijing University of Posts and Telecommunications Beijing China

Information Security Center Beijing University of Posts and Telecommunications Beijing China

School of Science, Beijing University of Posts and Telecommunications

Information Security Center, Beijing University of Posts and Telecommunications

北京交通大学计算机与信息技术学院计算机研究所

⁰