An Extension of Genetic Network Programming with Reinforcement Learning Using Actor-Critic.
Hiroyuki HatakeyamaShingo MabuKotaro HirasawaJinglu HuPublished in: IEEE Congress on Evolutionary Computation (2006)
Keyphrases
- actor critic
- reinforcement learning
- genetic network programming
- temporal difference
- policy gradient
- reinforcement learning algorithms
- approximate dynamic programming
- optimal control
- neuro fuzzy
- function approximation
- gradient method
- model free
- policy iteration
- state space
- learning algorithm
- association rule mining
- policy gradient methods
- control problems
- learning problems
- natural actor critic
- machine learning
- optimal policy
- association rules
- cost function
- dynamic programming
- rl algorithms
- step size
- multi agent systems
- multi agent
- least squares
- monte carlo
- supervised learning