Using the Hermite Regression Formula to Design a Neural Architecture with Automatic Learning of the "Hidden" Activation Functions.
Salvatore GaglioGiovanni PilatoFilippo SorbelloGiorgio VassalloPublished in: AI*IA (1999)
Keyphrases
- neural architecture
- activation function
- neural network
- neural nets
- feed forward
- artificial neural networks
- supervised learning
- back propagation
- learning algorithm
- prior knowledge
- multilayer perceptron
- basis functions
- learning rate
- fuzzy logic
- active learning
- radial basis function
- high dimensional
- support vector
- genetic algorithm