Empirical Evaluation of Activation Functions and Kernel Initializers on Deep Reinforcement Learning.
Sooyoung JangYoungsung SonPublished in: ICTC (2019)
Keyphrases
- empirical evaluation
- activation function
- reinforcement learning
- neural network
- artificial neural networks
- feed forward
- neural architecture
- hidden layer
- back propagation
- kernel function
- neural nets
- function approximation
- radial basis function
- learning rate
- input space
- kernel methods
- network architecture
- feature space
- hidden nodes
- multilayer perceptron
- state space
- training phase
- basis functions
- support vector
- learning algorithm
- dynamic programming
- high dimensional
- pattern recognition
- fuzzy neural network
- machine learning algorithms