Placing spline knots in neural networks using splines as activation functions.
Katerina HlaváckováMichel VerleysenPublished in: Neurocomputing (1997)
Keyphrases
- activation function
- b spline
- neural network
- basis functions
- cubic spline
- control points
- feed forward neural networks
- artificial neural networks
- back propagation
- feed forward
- hidden layer
- curve fitting
- hidden neurons
- neural nets
- multilayer perceptron
- neural architecture
- radial basis function
- linear combination
- network architecture
- feedforward neural networks
- hidden nodes
- learning rate
- connection weights
- least squares
- fault diagnosis
- training phase
- fuzzy neural network
- multi layer perceptron
- rbf neural network
- global exponential stability
- fuzzy logic
- training data
- genetic algorithm