Multi-Agent Reinforcement Learning and Chimpanzee Hunting

摘要：

The use of multi-agent reinforcement learning is growing because of its ability to scale in complexity and its lack of need for knowledge of the state and other agents. Chimpanzee hunting behavior is a suitable complex and interesting model for which multi-agent reinforcement learning is appropriate. Chimpanzee hunting strategies vary in both use and complexity and ultimately depend on the environment for which they are applied. Learning to use the varying strategies and learning when they are most effective is what this paper addresses and provides initial results and framework to build upon.

作者: Michael Z. Sauter Dongqing Shi Jerald D. Kralik

作者单位: Department of Psychological and Brain Science,Dartmouth College,6207 Moore Hall,Hanover,NH USA

会议类型: 国际会议

会议名称: 2009 IEEE International Conference on Robotics and Biomimetics(2009 IEEE 机器人与仿生技术国际会议 ROBIO 2009)

会议地点: 桂林

会议语种:英文

页码: 622-626

在线出版日期: 2009-12-19（万方平台首次上网日期，不代表论文的发表时间）