Trainable activation function with differentiable negative side and adaptable rectified point.
Kevin PratamaDae-Ki KangPublished in: Appl. Intell. (2021)
Keyphrases
- activation function
- neural network
- artificial neural networks
- feed forward
- hidden layer
- feedforward neural networks
- network size
- back propagation
- feed forward neural networks
- neural nets
- chaotic neural network
- learning rate
- objective function
- fuzzy neural network
- multilayer perceptron
- radial basis function
- hidden nodes
- input space
- pattern recognition