Learning Agents-Relations in Interactive Multiagent Dynamic Influence Diagrams

详细信息    查看全文

• 关键词：Interactive dynamic influence diagrams ; Relation learning ; Intelligent agents
• 刊名：Lecture Notes in Computer Science
• 出版年：2015
• 出版时间：2015
• 年：2015
• 卷：9145
• 期：1
• 页码：1-11
• 全文大小：1,088 KB
• 参考文献：1.Chandrasekaran, M., Doshi, P., Zeng, Y.: Approximate solutions of interactive dynamic influence diagrams using -behavioral equivalence. In: International Symposium on Artificial Intelligence and Mathematics (ISAIM) (2010)
  2.Doshi, P., Chandrasekaran, M., Zeng, Y.: Epsilon-subject equivalence of models for interactive dynamic influence diagrams. In: WIC/ACM/IEEE Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT) (2010)
  3.Doshi, P., Zeng, Y.: Improved approximation of interactive dynamic influence diagrams using discriminative model updates. In: AAMAS, pp. 907-14 (2009)
  4.Doshi, P., Zeng, Y., Chen, Q.: Graphical models for interactive pomdps: representations and solutions. J. Autonom. Agents Multi-Agent Syst. (JAAMAS) 18(3), 376-16 (2009)View Article
  5.Gal, K., Pfeffer, A.: Networks of influence diagrams: a formalism for representing agents-beliefs and decision-making processes. J. Artif. Intell. Res. 33, 109-47 (2008)MATH MathSciNet
  6.Gmytrasiewicz, P., Doshi, P.: A framework for sequential planning in multiagent settings. J. Artif. Intell. Res. (JAIR) 24, 49-9 (2005)MATH
  7.Howard, R.A., Matheson, J.E.: Influence diagrams. In: Readings on the Principles and Applications of Decision Analysis, pp. 721-62 (1984)
  8.Jensen, F.V.: Bayesian Networks and Decision Graphs. Information Science and Statistics. Springer, New York (2001)MATH View Article
  9.Kaelbling, L., Littman, M., Cassandra, A.: Planning and acting in partially observable stochastic domains. Artif. Intell. J. 101, 99-34 (1998)MATH MathSciNet View Article
  10.Koller, D., Milch, B.: Multi-agent influence diagrams for representing and solving games. In: International Joint Conference on Artificial Intelligence (IJCAI), pp. 1027-034 (2001)
  11.Pynadath, D., Marsella, S.: Minimal mental models. In: Twenty-Second Conference on Artificial Intelligence (AAAI), Vancouver, Canada, pp. 1038-044 (2007)
  12.Seuken, S., Zilberstein, S.: Formal models and algorithms for decentralized decision making under uncertainty. J. Autonom. Agents Multi-Agent Syst. (JAAMAS) 17(2), 190-50 (2008)View Article
  13.Tatman, J.A., Shachter, R.D.: Dynamic programming and influence diagrams. IEEE Trans. Syst. Man Cybern. 20(2), 365-79 (1990)MATH MathSciNet View Article
  14.Zeng, Y., Chen, Y., Doshi, P.: Approximating behavioral equivalence of models using top-k policy paths (extended abstract). In: International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), pp. 1229-230 (2011)
  15.Zeng, Y., Doshi, P.: Speeding up exact solutions of interactive influence diagrams using action equivalence. In: International Joint Conference on Artificial Intelligence (IJCAI) (2009)
  16.Zeng, Y., Doshi, P.: Exploiting model equivalences for solving interactive dynamic influence diagrams. J. Artif. Intell. Res. (JAIR) 43, 211-55 (2012)MATH MathSciNet
  17.Zeng, Y., Doshi, P., Chen, Q.: Approximate solutions of interactive dynamic influence diagrams using model clustering. In: Twenty Second Conference on Artificial Intelligence (AAAI), Vancouver, Canada, pp. 782-87 (2007)
  18.Zeng, Y., Doshi, P., Pan, Y., Mao, H., Chandrasekaran, M., Luo, J.: Utilizing partial policies for identifying equivalence of behavioral models. In: Twenty-Fifth Conference on Artificial Intelligence (AAAI), pp. 1083-088 (2011)
  19.Zeng, Y., Mao, H., Pan, Y., Luo, J.: Improved use of partial policies for identifying behavioral equivalence. In: Autonomous Agents and Multi-Agent Systems Conference (AAMAS), pp. 1015-022 (2012)
  20.Chen, Y., Doshi, P., Zeng, Y.: Iterative online planning in multiagent settings with limited model spaces and PAC guarantees. In: Autonomous Agents and Multi-Agent Systems Conference (AAMAS) (2015)
  
• 作者单位：Yinghui Pan (11) (12) (13) (14)
  Yifeng Zeng (12)
  Hua Mao (11) (12) (13) (14)
  
  11. School of Information Technology, Jiangxi University of Finance and Economics, Nanchang, China
  12. Department of Automation, Xiamen University, Xiamen, China
  13. School of Computing, Teesside University, Middlesbrough, UK
  14. College of Computer Science, Sichuan University, Chengdu, China
  
• 丛书名：Agents and Data Mining Interaction
• ISBN：978-3-319-20230-3
• 刊物类别：Computer Science
• 刊物主题：Artificial Intelligence and Robotics
  Computer Communication Networks
  Software Engineering
  Data Encryption
  Database Management
  Computation by Abstract Devices
  Algorithm Analysis and Problem Complexity
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1611-3349

文摘

Solving interactive multiagent decision making problems is a challenging task since it needs to model how agents interact over time. From individual agents-perspective, interactive dynamic influence diagrams?(I-DIDs) provide a general framework for sequential multiagent decision making in uncertain settings. Most of the current I-DID research focuses on the setting of \(n=2\) agents, which limits its general applications. This paper extends I-DIDs for \(n>2\) agents, which as expected increases the solution complexity due to the model space of other agents in the extended I-DIDs. We exploit data of agents-interactions to discover their relations thereby reducing the model complexity. We show preliminary results of the proposed techniques in one problem domain.