A Novel Multi-Agent Parallel-Critic Network Architecture for Cooperative-Competitive Reinforcement Learning.
Yu SunJun LaiLei CaoXi-liang ChenZhixiong XuYue XuPublished in: IEEE Access (2020)
Keyphrases
- network architecture
- multi agent
- cooperative
- reinforcement learning
- reinforcement learning algorithms
- function approximation
- artificial neural
- temporal difference
- actor critic
- neural network
- multi agent reinforcement learning
- neural network model
- multi agent environments
- multi agent systems
- multiagent systems
- multiagent reinforcement learning
- cooperating agents
- network design
- service integration
- activation function
- policy gradient
- learning algorithm
- dynamic programming
- state space
- cooperative agents
- connection weights
- parallel implementation
- multiple agents
- parallel processing
- approximate dynamic programming
- markov decision processes
- network layer
- game theory
- autonomous agents
- optimal policy
- model free
- learning agents
- reinforcement learning methods
- learning classifier systems
- shared memory