Browse > Article
http://dx.doi.org/10.5391/IJFIS.2007.7.4.285

Online Parameter Estimation and Convergence Property of Dynamic Bayesian Networks  

Cho, Hyun-Cheol (Dept. of Electrical Engineering, Dong-A University)
Fadali, M. Sami (Dept. of Electrical Engineering/260, University of Nevada)
Lee, Kwon-Soon (Dept. of Electrical Engineering, Dong-A University)
Publication Information
International Journal of Fuzzy Logic and Intelligent Systems / v.7, no.4, 2007 , pp. 285-294 More about this Journal
Abstract
In this paper, we investigate a novel online estimation algorithm for dynamic Bayesian network(DBN) parameters, given as conditional probabilities. We sequentially update the parameter adjustment rule based on observation data. We apply our algorithm to two well known representations of DBNs: to a first-order Markov Chain(MC) model and to a Hidden Markov Model(HMM). A sliding window allows efficient adaptive computation in real time. We also examine the stochastic convergence and stability of the learning algorithm.
Keywords
dynamic Bayesian networks; online parameter estimation; convergence property; sliding window; Markov chain;
Citations & Related Records
연도 인용수 순위
  • Reference
1 J. C. Stiller and G. Radons, 'Online estimation of hidden Markov models,' IEEE Signal Processing Letters, vol. 6, no. 8,pp.213-215, 1999   DOI   ScienceOn
2 D. Hernandez-Hernandez, S. I. Marcus, and P. J. Fard, 'Analysis of a risk-sensitive control problem for hidden Markov chains,' IEEE Trans. on Automatic Control, vol. 44, no. 5, pp. 1093-1100, 1999   DOI   ScienceOn
3 J. J. Ford and J. B. Moore, 'Adaptive estimation of HMM transition probabilities,' IEEE Trans. on Signal Processing, vol. 46, no. 5, pp. 1374-1385, 1998   DOI   ScienceOn
4 X. Huang, A. Acero, and H.-W. Hon, Spoken language processing, Prentice Hall, 2001
5 H. C. Cho, S. M. Fadali, and K. S. Lee, 'Estimation of non-Gaussian probability density by dynamic Bayesian networks,' Int. Conf on Control, Automation, and Systems, pp. 408-413, 2005
6 V. Krishnamurthy, J. B. Moore, and S.-H. Chung, 'Hidden Markov model signal processing in presence of unknown deterministic inferences,' IEEE Transaction on Automatic Control, vol. 38, no. 1, pp. 146-152,1993   DOI   ScienceOn
7 K. Murphy, 'Dynamic Bayesian networks: Representation, Inference and Learning.' Ph. D. Dissertation, UC Berkeley, 2002
8 P. Baldi and Y. Chauvin, 'Smooth on-line learning algorithm for hidden Markov models,' Neural Computation, vol. 6, no. 2, pp. 307-318, 1994   DOI   ScienceOn
9 J. M. Mendel, Lessons in estimation theory for signal processing, communications, and control, Prentice Hall, 1995
10 B. W. Silverman, Density Estimation for Statistics and Data Analysis, Chapman & Hall/CRC, 1986
11 I. Cohen and A. Bronstein, 'Adaptive online learning of Bayesian network parameters,' http://www.hpl.hp.com /techreports /2001/HPL-2001-156.pdf, 2001
12 W. J. Rugh, Linear system theory, Prentice Hall, 1996. 2004
13 E. Parzen, 'On estimation of a probability density function and mode,' Annals of Mathematical Statistics, vol. 33, no. 3, pp. 1065-1076, 1962   DOI
14 T. K. Moon, 'The expectation-maximization algorithm,' IEEE Signal Processing Magazine, vol. 13, no. 6, pp. 47-60, 1996   DOI   ScienceOn
15 S.-Z. Zhang, H. Yu, N-H. Yang, and X.-K Wang, 'An application of online learning algorithm for Bayesian network parameter,' Proc. of the Second International Conf. on Machine Learning and Cybernetics, pp. 153-156, 2003
16 J. Ying, T. Kirubarajan, K. R. Pattipati, A. Patterson-Hine, 'A hidden Markov model-based algorithm for fault diagnosis with partial and imperfect tests,' IEEE Trans. on Systems, Man and Cybernetics, vol. 30, no. 4, pp. 463-473, 2000   DOI   ScienceOn
17 R. J. Serfling, Approximation theorems of mathematical statistics, Wiley, 1980