Explaining slow convergence of EM in low noise linear mixtures

Kaare Brandt Petersen; Ole Winther

Explaining slow convergence of EM in low noise linear mixtures

Kaare Brandt Petersen, Ole Winther

Research output: Book/Report › Report › Research › peer-review

337 Downloads (Pure)

Abstract

This report conducts an investigation of the convergence properties of the EM algorithm used for linear mixture models. Since the linear mixture model is a rather general approach, the analysis is relevant for a wide range of models which to some degree are subsets of each other: Independent Component Analysis (ICA), probabilistic PCA, Factor Analysis (FA), Independent Factor Analysis (IFA) and Mean Field ICA.

Original language	English

Publication status	Published - 2005

Access to Document

Imm3444Final published version, 528 KB

http://www2.imm.dtu.dk/pubdb/p.php?3444

Fingerprint Dive into the research topics of 'Explaining slow convergence of EM in low noise linear mixtures'. Together they form a unique fingerprint.

Cite this

@book{72bf9d41ee714963b01b692a6ba7dbaa,

title = "Explaining slow convergence of EM in low noise linear mixtures",

abstract = "This report conducts an investigation of the convergence properties of the EM algorithm used for linear mixture models. Since the linear mixture model is a rather general approach, the analysis is relevant for a wide range of models which to some degree are subsets of each other: Independent Component Analysis (ICA), probabilistic PCA, Factor Analysis (FA), Independent Factor Analysis (IFA) and Mean Field ICA.",

author = "Petersen, {Kaare Brandt} and Ole Winther",

year = "2005",

language = "English",

}

TY - RPRT

T1 - Explaining slow convergence of EM in low noise linear mixtures

AU - Petersen, Kaare Brandt

AU - Winther, Ole

PY - 2005

Y1 - 2005

N2 - This report conducts an investigation of the convergence properties of the EM algorithm used for linear mixture models. Since the linear mixture model is a rather general approach, the analysis is relevant for a wide range of models which to some degree are subsets of each other: Independent Component Analysis (ICA), probabilistic PCA, Factor Analysis (FA), Independent Factor Analysis (IFA) and Mean Field ICA.

AB - This report conducts an investigation of the convergence properties of the EM algorithm used for linear mixture models. Since the linear mixture model is a rather general approach, the analysis is relevant for a wide range of models which to some degree are subsets of each other: Independent Component Analysis (ICA), probabilistic PCA, Factor Analysis (FA), Independent Factor Analysis (IFA) and Mean Field ICA.

M3 - Report

BT - Explaining slow convergence of EM in low noise linear mixtures

ER -