文章基本信息

标题：Cell Division Approach for Search Space in Reinforcement Learning
本地全文：下载
作者：Akira Notsu ; Hiroyuki Wada ; Katsuhiro Honda 等
期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2008
卷号：8
期号：6
页码：18-21
出版社：International Journal of Computer Science and Network Security
摘要：In this paper, we propose a state and action search space division algorithm during learning process like a cell division. This algorithm is designed for search domain reduction and heuristic space segmentation. In this method, the most activated space segment is divided into new two segments during its learning. Appropriate search domain reduction can minimize the learning time and enables us to recognize the evolutionary process. This segmentation method is also designed for social simulation models. In a way, social space segmentation, such as language systems and culture, will be revealed by multi-agent social simulation with our method.
关键词：Reinforcement Learning, Q-Learning, Cell Division, Agent Simulation