A comparison of linear ICA and local linear ICA for mutual information based feature ranking

Authors:
Tian Lan;Yonghong Huang;Deniz Erdogmus
Affiliations:
BME Department, OGI, Oregon Health & Science University, Portland, OR;CSEE Department, OGI, Oregon Health & Science University, Portland, OR;BME Department, OGI, Oregon Health & Science University, Portland, OR
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 4
Cited 0

Feature extraction by non parametric mutual information maximization

The Journal of Machine Learning Research
Blind source separation via generalized eigenvalue decomposition

The Journal of Machine Learning Research
ICA using spacings estimates of entropy

The Journal of Machine Learning Research
Input feature selection for classification problems

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Feature selection and dimensionality reduction is important for high dimensional signal processing and pattern recognition problems. Feature selection can be achieved by filter approach, in which certain criteria must be optimized. By using mutual information (MI) between feature vectors and class labels as the criterion, we proposed an ICA-MI framework for feature selection. In this paper, we will compare the linear ICA and local linear ICA for the accuracy of MI estimation, and study the bias-variance trade-off on feature projections and ranking.