Learning the structure of bayesian networks with constraint satisfaction

作者: David D. Jensen , Andrew S. Fast

DOI:

关键词:

摘要: A Bayesian network is graphical representation of the probabilistic relationships among set variables and can be used to encode expert knowledge about uncertain domains. The structure this model represents conditional independencies in data. networks are widely applicable, having been domains ranging from monitoring patients an emergency room predicting severity hailstorms. In thesis, I focus on problem learning Under certain assumptions, learned a represent causal data. Constraint-based algorithms for designed accurately identify distribution underlying data and, therefore, relationships. These use series hypothesis tests learn independence constraints model. When sample size limited, these prone errors. present comprehensive empirical evaluation constraint-based show that existing many false negative errors due running with low statistical power. Furthermore, analysis shows solutions fail reduce overall algorithms. I new inspired by constraint satisfaction able produce significant improvements structural accuracy. exploit interaction error. First, introduce algorithm based optimization sound limit, like algorithms, but guaranteed DAG. This learns models accuracy equivalent or better algorithms. Second, relaxation. Constraint relaxation combines different techniques likely incorrect, remove those consideration. combining produces significantly when compared demonstrating effectiveness approaches accurate networks.

参考文章(87)
Justus H. Piater, Paul R. Cohen, Xiaoqin Zhang, Michael Atighetchi, A Randomized ANDOVA Procedure for Comparing Performance Curves international conference on machine learning. pp. 430- 438 ,(1997)
Peter Spirtes, Christopher Meek, Learning Bayesian networks with discrete variables from data knowledge discovery and data mining. pp. 294- 299 ,(1995)
Carl M Kadie, David Maxwell Chickering, Robert Rounthwaite, David Heckerman, Chris Meek, Dependency networks for inference Journal of Machine Learning Research. ,(2000)
Liviu Badea, Determining the direction of causal influence in large probabilistic networks: a constraint-based approach european conference on artificial intelligence. pp. 263- 267 ,(2004)
Serafín Moral, Joaquín Abellán, Manuel Gómez-Olmedo, Some Variations on the PC Algorithm. probabilistic graphical models. pp. 1- 8 ,(2006)
Finn Verner Jensen, Allan Leck Jensen, MIDAS - an influence diagram for management of mildew in winter wheat uncertainty in artificial intelligence. pp. 349- 356 ,(1996)
D. E. Heckerman, E. J. Horvitz, B. N. Nathwani, Toward normative expert systems: Part I. The Pathfinder project. Methods of Information in Medicine. ,vol. 31, pp. 90- 105 ,(1992) , 10.1055/S-0038-1634867
David D. Jensen, Paul R. Cohen, Multiple Comparisons in Induction Algorithms Machine Learning. ,vol. 38, pp. 309- 338 ,(2000) , 10.1023/A:1007631014630
Ingo A. Beinlich, H. J. Suermondt, R. Martin Chavez, Gregory F. Cooper, The ALARM Monitoring System: A Case Study with two Probabilistic Inference Techniques for Belief Networks artificial intelligence in medicine in europe. pp. 247- 256 ,(1989) , 10.1007/978-3-642-93437-7_28