Estimating the Number of Hidden Neurons of the MLP Using Singular Value Decomposition and Principal Components Analysis: A Novel Approach.
José Daniel A. SantosGuilherme De A. BarretoCláudio M. S. MedeirosPublished in: SBRN (2010)
Keyphrases
- singular value decomposition
- principal components analysis
- number of hidden neurons
- multilayer perceptron
- dimensionality reduction
- hidden neurons
- hidden layer
- neural network
- artificial neural networks
- activation function
- learning rate
- principal component analysis
- covariance matrix
- linear discriminant analysis
- principal components
- radial basis function
- latent semantic indexing
- least squares
- back propagation
- singular values
- generalization ability
- neural network model
- feedforward neural networks
- dimension reduction
- support vector machine
- high dimensional
- multi layer perceptron
- low dimensional
- feature space
- feature selection
- feed forward
- high dimensional data
- data points
- pattern recognition
- feature extraction
- machine learning
- input variables
- bp neural network
- data fusion
- training data
- hidden nodes
- computer vision
- data sets