Differential Log Likelihood for Evaluating and Learning Gaussian Mixtures

Authors:
Marc M. Van Hulle
Affiliations:
K. U. Leuven, Laboratorium voor Neuro- en Psychofysiologie, B-3000 Leuven, Belgium
Venue:
Neural Computation
Year:
2006

Citing 5
Cited 1

Elements of information theory

Elements of information theory
Independent component analysis, a new concept?

Signal Processing - Special issue on higher order statistics
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Edgeworth Approximation of Multivariate Differential Entropy

Neural Computation
Maximum Likelihood Topographic Map Formation

Neural Computation

Learning nonlinear manifolds based on mixtures of localized linear manifolds under a self-organizing framework

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We introduce a new unbiased metric for assessing the quality of density estimation based on gaussian mixtures, called differential log likelihood. As an application, we determine the optimal smoothness and the optimal number of kernels in gaussian mixtures. Furthermore, we suggest a learning strategy for gaussian mixture density estimation and compare its performance with log likelihood maximization for a wide range of real-world data sets.