An Improvement Proposal for the Greedy Algorithm in Gaussian Mixture Estimation

• Andre Paim Lemos UFMG
• Antonio Pádua Braga UFMG

Abstract

This work proposes modifications on the stop criterion of the greedy algorithm for Gaussian Mixtures, in order to increase the accuracy in the search for the optimum number of mixture components. In this work, the stop criterion is modified in order to use a sampling multivariate normality test. The algorithm stops when all mixture components pass on the proposed test. The modified algorithm is compared with the original one, that uses parsimony criterion as stop criterion. Numerical simulation results suggest the accuracy improvement when the stop criterion proposed in this work is used.

References

Biernacki, C., Celeux, G., and Govaert, G. (1999). An improvement of the nec criterion for assessing the number of clusters in a mixture model. Non-Linear Anal., 20(3):267–272.

Bilmes, J. (1998). A gentle tutorial of the em algorithm and its application to parameter estimation for gaussian mixture and hidden markov models. Technical report.

Dempster, A. P., Laird, N. M., and Rubin, D. B. (1977). Maximum likelihood from incomplete data via the em algorithm. Journal of the Royal Statistical Society. Series B (Methodological), 39(1):1–38.

Duda, R., Hart, P., and Stork, D. (2001). Pattern classification. John Wiley & Sons, Inc., New York, NY, USA.

Figueiredo, M. A. T., Figueiredo, M. A. T., and Jain, A. K. (2000). Unsupervised learning of finite mixture models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24:381–396.

Li, L. and Ma, J. (2008). A BYY scale-incremental EM algorithm for Gaussian mixture learning. Applied Mathematics and Computation, 205(2, Sp. Iss. SI):832–840.

Mclachlan, G. J. (1987). On bootstrapping the likelihood ratio test stastistic for the number of components in a normal mixture. Applied Statistics, 36(3):318–324.

Papoulis, A. (1984). Probability, Random Variables, and Stochastic Processes. Mc-Graw Hill.

Ueda, N. and Nakano, R. (1998). Deterministic annealing em algorithm. Neural Netw., 11(2):271–282.

Verbeek, J. J., Vlassis, N., and Kröse, B. (2003). Efficient greedy learning of gaussian mixture models. Neural Comput., 15(2):469–485.

Ververidis, D. and Kotropoulos, C. (2005). Emotional speech classification using gaussian mixture models and the sequential floating forward selection algorithm. In Multimedia and Expo, 2005. ICME 2005. IEEE International Conference on, pages 1500–1503.

Ververidis, D. and Kotropoulos, C. (2008). Gaussian mixture modeling by exploiting the mahalanobis distance. Signal Processing, IEEE Transactions on, 56(7):2797–2811.