Communications for Statistical Applications and Methods

The Korean Statistical Society (한국통계학회)
권호 표지
DOI QR코드

DOI QR Code

Speedup of EM Algorithm by Binning Data for Normal Mixtures

혼합정규분포의 모수 추정에서 구간도수 EM 알고리즘의 실행 속도 개선

  • Published : 2008.01.31
Download PDF
⟨ Previous Next ⟩

Abstract

For a large data set the high computational cost of estimating the parameters of normal mixtures with the conventional EM algorithm is crucially impedimental in applying the algorithm to the areas requiring high speed computation such as real-time speech recognition. Simulations show that the binned EM algorithm, being compared to the standard one, significantly reduces the cost of computation without loss in accuracy of the final estimates.

혼합정규분포로부터 얻은 자료의 크기가 크면 EM 알고리즘으로 모수를 추정하는 경우 추정에 많은 시간이 걸리며 이는 실시간 음성인식 분야등에서는 적용이 어렵게 되는 문제가 발생한다. 대용량 자료를 구간도수로 요약하여 구간도수 EM 알고리즘을 적용하면 표준 EM 알고리즘에 비해 실행속도가 획기적으로 개선되며 더욱이 구간도수 EM 알고리즘에서의 추정치의 효율성이 표준 EM 알고리즘에 근접함을 시뮬레이션 실험을 통하여 보였다.

Keywords

References

  1. Cadez, I. V., McLachlan, G. J. and McLaren, C. E. (2002). Maximum likelihood estimation of mixture densities for binned and truncated multivariate data. Machine Learning, 47, 7-34 https://doi.org/10.1023/A:1013679611503
  2. 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, Ser. B, 39, 1-38
  3. Fu, Z., Yang, J., Hu, W. and Tan, T. (2004). Mixture clustering using multidi- mensional histogram for skin detection. In Proceedings of the 17th International Conference on Pattern Recognition, 4, 549-552
  4. McLachlan, G. J. and Jones, P. N. (1988). Fitting mixture models to grouped and truncated data via the EM algorithm. Biometrics, 44, 571-578 https://doi.org/10.2307/2531869
  5. McLachlan, G. J. and Krishnan, T. (1997). The EM Algorithm and Extensions. John Wiley & Sons, New York
  6. Rabiner, L. and Juang, B. (1993). Fundamentals of Speech Recognition, Prentice Hall, New Jersey
  7. Same, A., Ambroise, C. and Govaert, G. (2006). A classification EM algorithm for binned data. Computational Statistics & Data Analysis, 51, 466-480 https://doi.org/10.1016/j.csda.2005.08.009
  8. Stuttle, M. N. and Gales, M. J. F. (2001). A mixture of Gaussians front end for speech recognition. In Proceedings Eurospeech 2001
  9. Zolfaghari, P. and Robinson, T. (1996). Formant analysis using mixtures of Gaussians. In Proceedings ICSLP 96: Fourth International Conference on Spoken Language Processing
  10. Zolfaghari, P. and Robinson, T. (1997). A segmental formant vocoder based on linearly varying mixture of Gaussians. In Proceedings Eurospeech '97