Efficient pruning of multilayer perceptrons using a fuzzy sigmoid activation function.
Emilio Soria-OlivasJosé David Martín-GuerreroAntonio J. Serrano-LópezJavier Calpe-MaravillaJoan Vila-FrancésGustavo Camps-VallsPublished in: Neurocomputing (2006)
Keyphrases
- activation function
- multilayer perceptron
- hidden layer
- neural network
- artificial neural networks
- back propagation
- number of hidden neurons
- radial basis function
- feed forward
- feed forward neural networks
- fuzzy neural network
- neural nets
- hidden nodes
- feedforward neural networks
- backpropagation algorithm
- learning rate
- hidden neurons
- belief networks
- neural network model
- error function
- sigmoid function
- rbf neural network
- basis functions
- single hidden layer
- network architecture
- fuzzy sets
- rbf network
- recurrent neural networks
- fuzzy clustering
- training phase
- multi layer
- convergence speed
- bp neural network
- fuzzy logic