跳到主要导航 跳到搜索 跳到主要内容

Robust dirichlet process mixtures

  • University of Nottingham

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Non-parametric Dirichlet Process mixture (DPM) approaches for density estimation and clustering allow for automatic model selection. In this paper, we aim to develop robust DPM algorithm for clustering datasets with scatter objects, or outliers. In the developed mean-field variational inference algorithms, the auxiliary posterior distributions are factorized in a tree-structured form. In the experiments, we first show the advantage of the tree-structured factorization over the commonly-used full factorization. Then the performances of the robust DPM is evaluated using controlled experiment settings. Finally, the developed robust DPM is applied to biology datasets.

源语言英语
主期刊名Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
1556-1560
页数5
DOI
出版状态已出版 - 2011
已对外发布
活动2011 7th International Conference on Natural Computation, ICNC 2011 - Shanghai, 中国
期限: 26 7月 201128 7月 2011

丛书

姓名Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011
3

会议

会议2011 7th International Conference on Natural Computation, ICNC 2011
国家/地区中国
Shanghai
时期26/07/1128/07/11

学术指纹

探究 'Robust dirichlet process mixtures' 的科研主题。它们共同构成独一无二的学术指纹。

引用此