Learning Bayesian Network Parameters from Small Data Set: A Spatially Maximum a Posteriori Method

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 9505)

Abstract

To learn accurate BN parameters from small data set, combined with data, domain knowledge is often incorporated into the learning process as parameter constraints. Currently, most of the existing parameter learning methods take parameter learning problem as an exact optimization problem and regard the optimal solutions as the final parameters. However, due to the scarcity of data, objective functions constructed from the data, like likelihood function and entropy function, are not accurate. Therefore, parameters derived from the objective functions do not approach the true parameters well while some suboptimal parameters fit the true parameters better. Thus, searching more reasonable suboptimal parameters is a possible approach to learn better BN parameters. In this paper, we propose to visualize suboptimal parameters with parallel coordinate system and propose a Spatially Maximum a Posteriori (SMAP) method. Experimental results reveal that the proposed method outperforms most of the existing parameter learning methods.

Keywords

Bayesian Networks Parameter learning Small data set Convex optimization Linear programming 

Notes

Acknowledgements

This work is supported in part by National Nature Science Foundation of China (grant 61573285) and the Doctoral Fund of Ministry of Education of China (grant 20116102110026).

References

  1. 1.
    Pearl, J.: Probabilistic Reasoning in Intelligent Systems. Morgan Kaufmann, Burlington (1988)MATHGoogle Scholar
  2. 2.
    Tamda, Y., Imoto, S., Araki, H.: Estimating genome-wide gene networks using nonparametric Bayesian network models on massively parallel computers. IEEE Trans. Comput. Biol. Bioinform. 3, 683–697 (2011)CrossRefGoogle Scholar
  3. 3.
    Ibrahim, W., Beiu, V.: Using Bayesian networks to accurately calculate the reliability of complementary metal oxide semiconductor gates. IEEE Trans. Reliab. 60, 538–549 (2011)CrossRefGoogle Scholar
  4. 4.
    Infantes, G., Ghallab, M., Ingrand, F.: Learning the behavior model of a robot. Auton. Robot. 30, 157–177 (2011)CrossRefGoogle Scholar
  5. 5.
    Steven, M., Ann, N., Kevin, K.: Anomaly detection in vessel tracks using Bayesian networks. Int. J. Approximate Reasoning 55, 84–98 (2014)CrossRefGoogle Scholar
  6. 6.
    Neil, W., Mahmood, R., Kevin, K.: Detection and classification of non-stationary transient signals using sparse approximations and Bayesian networks. IEEE/ACM Trans. Audio Speech Lang. Process. 22, 1750–1764 (2014)CrossRefGoogle Scholar
  7. 7.
    Dries, L., Steven, B., Rob, D., Guy, E., Joris, A.: A review of Bayesian belief networks in ecosystem service modeling. Environ. Model. Softw. 46, 1–11 (2013)CrossRefGoogle Scholar
  8. 8.
    Redner, R., Walker, H.: Mixture densities, maximum likelihood and the EM algorithm. SIAM Rev. 26, 195–239 (1984)MathSciNetCrossRefGoogle Scholar
  9. 9.
    Helsper, E., Gaag, L., Groenendal, F.: Designing a procedure for the acquisition of probability constraints for Bayesian networks. In: Proceedings of the Fourteenth Conference on Engineering Knowledge in the Age of the Semantic Web, pp. 280–292 (2004)CrossRefGoogle Scholar
  10. 10.
    Wittig, F., Jameson, A.: Exploiting qualitative knowledge in the learning of conditional probabilities of Bayesian networks. In: Proceedings of the Sixteenth International Conference on Uncertainty in Artificial Intelligence, pp. 644–652 (2000)Google Scholar
  11. 11.
    Altendorf, E., Restificar, A., Dietterich, T.: Learning from sparse data by exploiting monotonicity constraints. In: Proceedings of the Twenty First International Conference on Uncertainty in Artificial Intelligence, pp. 18–26 (2005)Google Scholar
  12. 12.
    Feelders, A., Gaag, L.: Learning Bayesian networks parameters under order constraints. Int. J. Approximate Reasoning 42, 37–53 (2006)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Isozaki, T., Kato, N., Ueno, M.: Data temperature in minimum free energies for parameter learning of Bayesian networks. Int. J. Artif. Intell. Tools 18, 653–671 (2009)CrossRefGoogle Scholar
  14. 14.
    Niculescu, R., Mitchell, T., Rao, B.: Bayesian network learning with parameter constraints. J. Mach. Learn. Res. 7, 1357–1383 (2006)MathSciNetMATHGoogle Scholar
  15. 15.
    Campos, C., Yan, T., Qiang, J.: Constrained maximum likelihood learning of Bayesian networks for facial action recognition. In: Proceedings of the Tenth European Conference on Computer Vision, pp. 168–181 (2008)Google Scholar
  16. 16.
    Campos, C., Qiang, J.: Improving Bayesian network parameter learning using constraints. In: Proceedings of the Nineteenth International Conference on Pattern Recognition, pp. 1–4 (2008)Google Scholar
  17. 17.
    Rui, C., Shoemaker, R., Wei, W.: A novel knowledge-driven systems biology approach for phenotype prediction upon genetic intervention. IEEE Trans. Comput. Biol. Bioinform. 1, 1170–1181 (2011)Google Scholar
  18. 18.
    Zhi-gao, G., Xiao-guang, G., Ruo-hai, D.: Learning Bayesian network parameters under dual constraints from small data set. Acta Automatica Sinica 40, 1509–1516 (2014)Google Scholar
  19. 19.
    Yun, Z., Fenton, N., Neil, M.: Bayesian network approach to multinomial parameter learning using data and expert judgments. Int. J. Approximate Reasoning 55, 1252–1268 (2014)MathSciNetCrossRefGoogle Scholar
  20. 20.
    Dasgupta, S.: The sample complexity of learning fixed-structure Bayesian networks. Mach. Learn. 29, 165–180 (1997)CrossRefGoogle Scholar
  21. 21.
    Boyd, S.: CVX: Matlab software for disciplined convex programming (2015). http://cvxr.com/cvx
  22. 22.
    Kullback, S., Leibler, R.: On information and sufficiency. Ann. Math. Stat. 22, 79–86 (1951)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Open Access This chapter is distributed under the terms of the Creative Commons Attribution Noncommercial License, which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Authors and Affiliations

  • Zhi-gao Guo
    • 1
  • Xiao-guang Gao
    • 1
  • Ruo-hai Di
    • 1
  • Yu Yang
    • 1
  1. 1.Department of System EngineeringNorthwestern Polytechnical UniversityXianChina

Personalised recommendations