Comments on 'Efficient Training Algorithms for HMM's Using Incremental Estimation'

Comments on 'Efficient Training Algorithms for HMM's Using Incremental Estimation'” by W. Byrne and A. Gunawardana. IEEE Transactions on Speech and Audio Processing, vol. 8, no. 6, Nov. 2000, pp. 751-754 (4 pages).

Abstract

“Efficient Training Algorithms for HMM's using Incremental Estimation” investigates EM procedures that increase training speed. The authors' claim that these are GEM procedures is incorrect. We discuss why this is so, provide an example of non-monotonic convergence to a local maximum in likelihood, and outline conditions that guarantee such convergence.

BibTeX entry:

@article{byrne00:_comments,
   author = {W. Byrne and A. Gunawardana},
   title = {Comments on '{Efficient} Training Algorithms for {HMM}'s Using
	Incremental Estimation'},
   journal = {IEEE Transactions on Speech and Audio Processing},
   volume = {8},
   number = {6},
   pages = {751--754 (4 pages)},
   month = nov,
   year = {2000},
   url = {http://dx.doi.org/10.1109/89.876315}
}

Back to Bill Byrne publications.