Cooperative Behavior Rule Acquisition for Multi-Agent Systems by Machine Learning [chapter]

Mengchun Xie
2011 Advances in Reinforcement Learning  
Advances in Reinforcement Learning 82 Second, we construct the learning agent using the Q-learning which is a representative technique of reinforcement learning. Q-learning is a method to let an agent learn from delayed reward and punishment. It is designed to find a policy that maximizes for all states. The decision policy is represented by a function. The action vale function is shares among agents. The third, we concentrate on an application of Multi-agent systems to disaster relief using
more » ... ter relief using Q-learning. We constructed a simplified disaster relief multi-agent system and acquired action rules by Q-learning. We then observe how the autonomous agents obtain their action rules and examined the influence of the learning situations on the system. Moreover, we discuss how the system was influenced by learning situation and the view information of the agent.
doi:10.5772/13280 fatcat:l2ify64kxrfdjn7ypojbdkppdi