Neural Networks
Testing for the number of components in a mixture of normal distributions using moment estimators
Computational Statistics & Data Analysis
Bayesian forecasting and dynamic models (2nd ed.)
Bayesian forecasting and dynamic models (2nd ed.)
Pattern Recognition and Neural Networks
Pattern Recognition and Neural Networks
The Probabilistic Growing Cell Structures Algorithm
ICANN '97 Proceedings of the 7th International Conference on Artificial Neural Networks
Self-organizing neural networks based on gaussian mixture model for pdf estimation and pattern classification
Probability Density Estimation Using Adaptive Activation Function Neurons
Neural Processing Letters
Model Validation for Model Selection
ICAPR '01 Proceedings of the Second International Conference on Advances in Pattern Recognition
An incremental neural network with a reduced architecture
Neural Networks
Hi-index | 0.00 |
We address the problem of estimating an unknown probabilitydensity function from a sequence of input samples. Weapproximate the input density with a weighted mixtureof a finite number of Gaussian kernels whose parameters andweights we estimate iteratively from the input samples using the MaximumLikelihood (ML) procedure.In order to decide on the correct total number ofkernels we employ simple statistical tests involving the mean, variance,and the kurtosis, or fourth moment, of aparticular kernel. We demonstrate the validity of ourmethod in handling both pattern classification (stationary) and time series(nonstationary) problems.