N. Ueda, R. Nakano
{"title":"EM algorithm with split and merge operations for mixture models","authors":"N. Ueda, R. Nakano","doi":"10.1002/(SICI)1520-684X(200005)31:5%3C1::AID-SCJ1%3E3.0.CO;2-G","DOIUrl":null,"url":null,"abstract":"The maximum-likelihood estimate of a mixture model is usually found by using the EM algorithm. However, the EM algorithm suffers from the local-optimum problem and therefore we cannot obtain the potential performance of mixture models in practice. In the case of mixture models, local maxima often involve having too many components of a mixture model in one part of the space and too few in another, widely separated part of the space. To escape from such configurations, we repeatedly perform simultaneous split and merge operations using a new criterion for efficiently selecting the split and merge candidates. We apply the proposed algorithm to the training of Gaussian mixtures and the dimensionality reduction based on a mixture of factor analyzers using synthetic and real data and show the effectiveness of using the split and merge operations to improve the likelihood both of the training data and of reserved test data. © 2000 Scripta Technica, Syst Comp Jpn, 31(5): 1–11, 2000","PeriodicalId":0,"journal":{"name":"","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2000-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"20","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1002/(SICI)1520-684X(200005)31:5%3C1::AID-SCJ1%3E3.0.CO;2-G","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 20
混合模型的分割和合并的EM算法
混合模型的最大似然估计通常采用EM算法。然而,电磁算法存在局部最优问题,因此在实际应用中无法获得混合模型的潜在性能。在混合模型的情况下,局部极大值通常涉及在空间的一部分中混合模型的组件太多,而在空间的另一部分中混合模型的组件太少。为了避免这样的配置,我们使用一个新的标准来重复执行拆分和合并操作,以有效地选择拆分和合并候选项。我们将提出的算法应用于高斯混合训练和基于混合因子分析的降维,使用合成数据和真实数据,并展示了使用分割和合并操作来提高训练数据和保留测试数据的似然性的有效性。©2000 Scripta Technica, system Comp, 31(5): 1 - 11,2000
本文章由计算机程序翻译,如有差异,请以英文原文为准。