Variational learning for finite Beta-Liouville mixture models |
| |
Authors: | LAI Yu-ping ZHOU Ya-jian PING Yuan GUO Yu-cui YANG Yi-xian Information Security Center,Beijing University of Posts Telecommunications |
| |
Affiliation: | LAI Yu-ping;ZHOU Ya-jian;PING Yuan;GUO Yu-cui;YANG Yi-xian;Information Security Center, Beijing University of Posts and Telecommunications;Department of Computer Science and Technology, Xuchang University;School of Science, Beijing University of Posts and Telecommunications; |
| |
Abstract: | In the article, an improved variational inference (VI) framework for learning finite Beta-Liouville mixture models (BLM) is proposed for proportional data classification and clustering. Within the VI framework, some non-linear approximation techniques are adopted to obtain the approximated variational object functions. Analytical solutions are obtained for the variational posterior distributions. Compared to the expectation maximization (EM) algorithm which is commonly used for learning mixture models, underfitting and overfitting events can be prevented. Furthermore, parameters and complexity of the mixture model (model order) can be estimated simultaneously. Experiment shows that both synthetic and real-world data sets are to demonstrate the feasibility and advantages of the proposed method. |
| |
Keywords: | variational inference model selection factorized approximation Beta-Liouville distribution mixing modeling |
本文献已被 CNKI 维普 ScienceDirect 等数据库收录! |