Simplifying computation of dynamic influence diagrams
暂无分享,去创建一个
Influence diagrams (EDs) are based on Bayesian networks (BNs) and decision theory, and they are powerful tools for representing and processing problems of agents. An approach of decomposition and incorporation is developed to solve problems of multi-agent system based on influence diagrams and dynamic Bayesian networks (DBNs) that is intractable for exact calculation. In additional, we discuss realizing decision of dynamic influence diagrams (DIDs) and reducing computation of decision problems.
[1] Alex Pentland,et al. A Bayesian Computer Vision System for Modeling Human Interactions , 1999, IEEE Trans. Pattern Anal. Mach. Intell..
[2] Zoubin Ghahramani,et al. An Introduction to Hidden Markov Models and Bayesian Networks , 2001, Int. J. Pattern Recognit. Artif. Intell..
[3] Svetha Venkatesh,et al. Learning other Agents' Preferences in Multiagent Negotiation , 1996, AAAI/IAAI, Vol. 1.
[4] Anders L. Madsen,et al. LAZY Propagation: A Junction Tree Inference Algorithm Based on Lazy Evaluation , 1999, Artif. Intell..