Sigmoid-weighted linear units for neural network function approximation in reinforcement learning.
Stefan ElfwingEiji UchibeKenji DoyaPublished in: Neural Networks (2018)
Keyphrases
- reinforcement learning methods
- function approximation
- function approximators
- reinforcement learning
- neural network
- mountain car
- reinforcement learning algorithms
- temporal difference learning algorithms
- temporal difference learning
- reinforcement learning problems
- model free
- radial basis function
- learning tasks
- temporal difference
- mackey glass
- activation function
- state space
- hidden layer
- supervised learning
- temporal difference methods
- learning algorithm
- back propagation
- machine learning
- dynamic programming
- active learning
- feature vectors
- feed forward
- artificial neural networks
- continuous state
- policy search
- markov decision processes
- transfer learning