Properties of the Hermite Activation Functions in a Neural Approximation Scheme.
Bartlomiej BeliczynskiPublished in: ICANNGA (2) (2007)
Keyphrases
- activation function
- neural network
- network architecture
- neural architecture
- continuous functions
- artificial neural networks
- feed forward
- hidden layer
- back propagation
- feed forward neural networks
- neural nets
- multilayer perceptron
- neural network model
- basis functions
- learning rate
- high dimensional data
- fuzzy neural network
- probabilistic model
- genetic algorithm
- dynamic programming
- artificial intelligence
- hidden nodes