Newton method for the ICA mixture model

13 years 11 months ago

Download sccn.ucsd.edu

We derive an asymptotic Newton algorithm for Quasi Maximum Likelihood estimation of the ICA mixture model, using the ordinary gradient and Hessian. The probabilistic mixture framework can accommodate non-stationary environments and arbitrary source densities. We prove asymptotic stability when the source models match the true sources. An application to EEG segmentation is given.

Jason A. Palmer, Scott Makeig, Kenneth Kreutz-Delg

Real-time Traffic

ICA Mixture Model | ICASSP 2008 | Probabilistic Mixture Framework | Quasi Maximum Likelihood | Signal Processing |

claim paper

» Learned parametric mixture based ICA algorithm

» Probabilistic PCA and ICA Subspace Mixture Models for Image Segmentation

» Video Event Detection using ICA Mixture Hidden Markov Models

» Unsupervised image classification segmentation and enhancement using ICA mixture models

» An ICA Mixture Hidden Markov Model for Video Content Analysis

» A novel vector quantizationbased video summarization method using independent component an...

» ICAbased Clustering of Genes from Microarray Expression Data

» Blind Maximum Likelihood Separation of a LinearQuadratic Mixture

Post Info
More Details (n/a)

Added	30 May 2010
Updated	30 May 2010
Type	Conference
Year	2008
Where	ICASSP
Authors	Jason A. Palmer, Scott Makeig, Kenneth Kreutz-Delgado, Bhaskar D. Rao

Comments (0)

Sciweavers

Newton method for the ICA mixture model

ICA Mixture Model | ICASSP 2008 | Probabilistic Mixture Framework | Quasi Maximum Likelihood | Signal Processing |

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers