Multi-Agent Actor-Critic with Generative Cooperative Policy Network.
Heechang RyuHayong ShinJinkyoo ParkPublished in: CoRR (2018)
Keyphrases
- cooperative
- multi agent
- actor critic
- reinforcement learning
- policy gradient
- multi agent systems
- temporal difference
- approximate dynamic programming
- neuro fuzzy
- optimal control
- function approximation
- gradient method
- partially observable markov decision processes
- single agent
- machine learning
- reinforcement learning algorithms
- optimal policy
- average reward
- sufficient conditions
- neural network
- natural actor critic