Unsupervised Learning for Finite Mixture Models Based on BIC Criterion and Gibbs Sampling

LIU Wei-feng; YANG Ai-lan

您当前的位置：

首页 >

文章列表页 >

Unsupervised Learning for Finite Mixture Models Based on BIC Criterion and Gibbs Sampling

更新时间：2025-07-16

- Unsupervised Learning for Finite Mixture Models Based on BIC Criterion and Gibbs Sampling
- Acta Electronica Sinica Vol. 39, Issue 3A, Pages: 134-139(2011)
- 作者机构：
  
  杭州电子科技大学自动化学院,浙江,杭州,310018
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP18
- Published：2011
- 稿件说明：
移动端阅览
LIU Wei-feng, YANG Ai-lan. Unsupervised Learning for Finite Mixture Models Based on BIC Criterion and Gibbs Sampling[J]. Acta Electronica Sinica, 2011, 39(3A): 134-139.
DOI：

LIU Wei-feng, YANG Ai-lan. Unsupervised Learning for Finite Mixture Models Based on BIC Criterion and Gibbs Sampling[J]. Acta Electronica Sinica, 2011, 39(3A): 134-139. DOI：

摘要

针对有限混合模型无监督学习算法分布元个数未知

本文提出了一种基于BIC准则和Gibbs采样的无监督学习算法

通过Gibbs采样算法对混合模型的参数进行估计

进一步计算观测数据的Bayes信息准则(BIC)指标

确定混合分布元个数.仿真实验以高斯混合分布为例

分别利用具有不同个数的分布模型拟合观测数据

分析表明

该算法能够很好地学习混合高斯分布参数及个数.

Abstract

An unsupervised learning algorithm for finite mixture models (FMM) by using the BIC criterion and the Gibbs sampling is proposed.The FMM parameters are estimated by the Gibbs sampling algorithm.The number of the models is further given by calculating the BIC criterion.In the final simulations

we propose two examples of Gaussian mixture models

where the FMM with different number of components is adopted to fit observation data.The final result shows that the proposed algorithm can effectively estimate the parameters and the number of the components.

关键词

Keywords

references

Views

3234

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

An Algorithm Based on Modified Momentum Using Restricted Boltzmann Machine

Robust Unsupervised Feature Selection with Double Fuzzy Learning

Feature Anomaly Detection and Pseudo-Label Regression for Adversarial Domain Adaptation

Unsupervised Monocular Depth Estimation Based on Scale Clue Enhancement

Anomaly Detection with Dual-Channel Heterogeneous Graph Neural Network Based on Hypersphere Dual Learning

Related Author

SHEN Hui-hui

LIU Guo-wu

FU Li-hua

LIU Zhi-hui

LI Hong-wei

CAO Chao

ZHAO Zhi-xiang

SHI Shu-guang

Related Institution

中国地质大学数理学院

湖北经济学院信息管理与统计学院

中国地质大学(武汉)地球内部多尺度成像湖北省重点实验室

Tung Sah Institute of Micro-Nano Science and Technology, Xiamen University

School of Ocean Information Engineering, Jimei University

⁰