Discrete dynamic BN parameter learning under small sample and incomplete information

被引：4
作者：
Ren, Jia [1 ]
Gao, Xiao-Guang [2 ]
Bai, Yong [1 ]
机构：
[1] College of Information Science and Technology, Hainan University
[2] College of Electronic Engineering, Northwestern Polytechnical University
来源：
Xi Tong Gong Cheng Yu Dian Zi Ji Shu/Systems Engineering and Electronics | 2012年 / 34卷 / 08期
关键词：
Constraint recursion learning; Discrete dynamic Bayesian network; Incomplete information; Parameter learning;
D O I：
10.3969/j.issn.1001-506X.2012.08.33
中图分类号：
学科分类号：
摘要：
Aiming at the discrete dynamic Bayesian network parameter learning under the situation of small sample and incomplete information, a constraint recursion learning algorithm is presented. The forward algorithm is used to establish a parameter recursion estimation model of discrete dynamic Bayesian network with hidden variables. A prior parameter constraint model with uniform distribution is established with the present network parameters as variables. Then the approximate Beta distribution could be acquired through the optimization algorithm. Finally, the distribution of prior parameter knowledge could be used in the above model of recursive estimation to finish the parameter learning process. The method is applied to the unmanned aerial vehicle dynamic model of threat assessment. The results show the effectiveness and accuracy of the proposed algorithm.
引用
收藏
页码：1723 / 1728
页数：5
相关论文
共 16 条
  • [1] Meng D., Sivakumar K., Kargupta H., Privacy-sensitive Bayesian network parameter learning, Proc. of the Fourth IEEE International Conference on Data Mining, pp. 487-490, (2004)
  • [2] Friedman N., The Bayesian structural EM algorithm, Proc. of the Fourteenth Annual Conference on Uncertainty in Artificial Intelligence, pp. 125-133, (1998)
  • [3] Lamine F.B., Kalti K., Mahjoub M.A., The threshold EM algorithm for parameter learning in Bayesian network with incomplete data, International Journal of Advanced Computer Science and Applications, 2, 7, pp. 86-91, (2011)
  • [4] Ramoni M., Sebastiani P., Robust learning with missing data, Machine Learning Archive, 45, 2, pp. 147-170, (2001)
  • [5] Liao W.H., Ji Q., Learning Bayesian network parameters under incomplete data with domain knowledge, Journal Pattern Recognition, 42, 11, pp. 3046-3056, (2009)
  • [6] Feelders A., A new parameter learning method for bayesian networks with qualitative influences, Proc. of the 23rd Conference on Uncertainty in Artificial Intelligence, pp. 117-124, (2007)
  • [7] Altendorf E., Restificar A.C., Dietterich T.G., Learning from sparse data by exploiting monotonicity constraints, Proc. of the 21st Conference in Uncertainty in Artificial Intelligence, pp. 18-26, (2005)
  • [8] Campos C.P., Tong Y., Ji Q., Constrained maximum likelihood learning of Bayesian networks for facial action recognition, Proc. of the 10th European Conference on Computer Vision Part III, pp. 168-181, (2008)
  • [9] Campos C.P., Zeng Z., Ji Q., Structure learning of Bayesian networks using constraints, Proc. of the 26th Annual International Conference on Machine Learning, pp. 113-120, (2009)
  • [10] Lee S.H., Lim S., Cho S.B., Parameter learning in Bayesian network using semantic constraints of conversational feedback, Proc. of the 11th Pacific Rim International Conference on Trends in Artificial Intelligence, pp. 467-476, (2010)
12