Periodic Activation Functions Induce Stationarity.
Lassi MeronenMartin TrappArno SolinPublished in: NeurIPS (2021)
Keyphrases
- activation function
- neural network
- artificial neural networks
- feed forward
- complex valued
- feed forward neural networks
- non stationary
- back propagation
- hidden layer
- neural nets
- neural architecture
- network architecture
- multilayer perceptron
- radial basis function
- learning rate
- hidden nodes
- autoregressive
- fuzzy neural network
- rbf neural network
- basis functions
- neural network model
- training phase
- continuous functions
- input space
- fuzzy logic
- decision making
- high dimensional data
- dynamic programming