Reinforcement learning of dynamic behavior by using recurrent neural networks.
Ahmet OnatHajime KitaYoshikazu NishikawaPublished in: Artif. Life Robotics (1997)
Keyphrases
- recurrent neural networks
- dynamic behavior
- reinforcement learning
- neural network
- feed forward
- feedforward neural networks
- recurrent networks
- reservoir computing
- function approximation
- artificial neural networks
- echo state networks
- complex valued
- neural model
- reinforcement learning algorithms
- hebbian learning
- optimal policy
- speed control
- learning algorithm
- cascade correlation
- temporal difference
- long short term memory
- model free
- markov decision processes
- state space
- dynamic networks
- fuzzy logic
- multi agent
- real time
- fuzzy systems
- optimal control
- machine learning
- data mining