Which Activation Function Works Best for Training Artificial Pancreas: Empirical Fact and Its Theoretical Explanation.
Lehel Dénes-FazakasLászló SzilágyiGyörgy EignerOlga KoshelevaMartine CeberioVladik KreinovichPublished in: SSCI (2023)
Keyphrases
- activation function
- feed forward neural networks
- feedforward neural networks
- backpropagation algorithm
- learning speed
- neural network
- hidden layer
- training phase
- single hidden layer
- artificial neural networks
- back propagation
- feed forward
- training algorithm
- multi layer perceptron
- learning rate
- neural nets
- multilayer perceptron
- output layer
- training set
- network architecture
- extreme learning machine
- hidden nodes
- radial basis function
- supervised learning
- training process
- basis functions
- sigmoid function
- chaotic neural network
- autoregressive
- multi layer
- rbf neural network
- recurrent neural networks
- input output
- non stationary
- machine learning