For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Component Reduction for Gaussian Mixture Models
Kumiko MAEBASHI Nobuo SUEMATSU Akira HAYASHI
IEICE TRANSACTIONS on Information and Systems
Publication Date: 2008/12/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: PAPER
Category: Pattern Recognition
mixture model, EM-algorithm, maximum likelihood, hierarchical clustering,
Full Text: PDF>>
The mixture modeling framework is widely used in many applications. In this paper, we propose a component reduction technique, that collapses a Gaussian mixture model into a Gaussian mixture with fewer components. The EM (Expectation-Maximization) algorithm is usually used to fit a mixture model to data. Our algorithm is derived by extending mixture model learning using the EM-algorithm. In this extension, a difficulty arises from the fact that some crucial quantities cannot be evaluated analytically. We overcome this difficulty by introducing an effective approximation. The effectiveness of our algorithm is demonstrated by applying it to a simple synthetic component reduction task and a phoneme clustering problem.