全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种交互式动态影响图的改进算法

, PP. 506-513

Keywords: Agent建模,交互式动态影响图,动态决策,ε-行为等价,信度-行为图

Full-Text   Cite this paper   Add to My Lib

Abstract:

交互式动态影响图(I-DIDs)是基于概率图形理论的多智能体动态交互决策的图模型。为缓解该模型状态空间随时间片增加呈指数级增长的趋势,文中基于行为等价的基本思想压缩状态空间,提出构建Epsilon行为等价类的方法:利用有向无环图表示其它Agent可能的信度和行为,把信度在空间上接近的模型聚为一类,实现自顶向下合并行为等价模型。该过程避免求解状态空间中的所有候选模型,节省了存储空间和计算时间。模型实例上的仿真结果显示了该算法的有效性。

References

[1]  Tatman J A,Shachter R D.Dynamic Programming and Influence Diagrams.IEEE Trans on Systems,Man and Cybernetics,1990,20: 365-379
[2]  Yao Hongliang,Wang Hao,Zhang Yousheng,et al.Multi-Agent Dynamic Influence Diagrams and Its Approximation of Probability Distribution.Pattern Recognition and Artificial Intelligence,2007,20(4): 521-532 (in Chinese)(姚宏亮,王 浩,张佑生,等.多Agent动态影响图及其概率分布的近似方法.模式识别与人工智能,2007,20(4): 521-532)
[3]  Yao Hongliang,Wang Hao,Wang Ronggui,et al.Approximate Computation of Multi-Agent Dynamic Influence Diagrams.Journal of Computer Research and Development,2008,45(3): 487-495 (in Chinese)(姚宏亮,王 浩,汪荣贵,等.多Agent动态影响图的近似计算方法.计算机研究与发展,2008,45(3): 487-495)
[4]  Gmytrasiewicz P J,Doshi P.A Framework for Sequential Planning in Multi-Agent Settings.Journal of Artificial Intelligence Research,2005,24(1): 49-79
[5]  Doshi P,Zeng Y F,Chen Q Y.Graphical Models for Interactive POMDPs: Representation and Solutions.Journal of Autonomous Agents and Multi-Agent Systems,2009,18(3): 376-416
[6]  Polich K,Gmytrasiewicz P J.Interactive Dynamic Influence Diagrams // Proc of the 6th International Joint Conference on Autonomous Agents and Multiagent Systems,New York,USA: ACM Press,2007: 147-149
[7]  Zeng Y F,Doshi P,Chen Q Y.Approximate Solutions of Interactive Dynamic Influence Diagrams Using Model Clustering // Proc of the 22nd International Conference on Association for the Advancement of Artificial Intelligence.Vancouver,Canada: AAAI Press,2007: 782-787
[8]  Zeng Y F,Doshi P.Speeding up Exact Solutions of Interactive Dynamic Influence Diagrams Using Action Equivalence // Proc of the 21st International Joint Conference on Artificial Intelligence.Pasadena,USA,2009: 1996-2001
[9]  Doshi P,Zeng Y F.Improved Approximation of Interactive Dynamic Influence Diagrams Using Discriminative Model Updates // Proc of the 8th International Conference on Autonomous Agents and Multi-Agent Systems.Budapest,Hungray,2009: 907-914
[10]  Smallwood R D,Sondik E J.The Optimal Control of Partially Observable Markov Decision Processes over a Finite Horizon.Operations Research,1973,21(5): 1071-1088
[11]  Pynadath D V,Marsella S C.Minimal Mental Models // Proc of the 22nd International Conference on Association for the Advancement of Artificial Intelligence.Vancouver,Canada,2007: 1038-1044
[12]  Geng S Y,Qun W L.Discrete Mathematics.Beijing: Higher Education Press,1998(耿素云,屈婉玲.离散数学.北京:高等教育出版社,1998)

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133