Actor-Critic Reinforcement Learning with Neural Networks in Continuous Games.
Gabriel LeuenbergerMarco A. WieringPublished in: ICAART (2) (2018)
Keyphrases
- actor critic
- reinforcement learning
- neural network
- policy gradient
- temporal difference
- optimal control
- approximate dynamic programming
- function approximation
- reinforcement learning algorithms
- neuro fuzzy
- gradient method
- action space
- function approximators
- stochastic games
- average reward
- control problems
- state space
- artificial neural networks
- fuzzy logic
- dynamic programming
- rl algorithms
- multi agent
- model free
- policy iteration
- policy gradient methods
- evaluation function
- markov decision processes
- learning algorithm
- genetic algorithm
- game theoretic
- feed forward
- learning problems
- transfer learning
- cost function
- machine learning
- action selection
- learning capabilities
- recurrent neural networks
- imperfect information
- optimal policy
- control system