DMCA
Large Margin Training for Hidden Markov Models with Partially Observed States Trinh-Minh-Tri Do
| Citations: | 42 - 4 self |
BibTeX
@MISC{Artières_largemargin,
author = {Thierry Artières},
title = {Large Margin Training for Hidden Markov Models with Partially Observed States Trinh-Minh-Tri Do},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Large margin learning of Continuous Density HMMs with a partially labeled dataset has been extensively studied in the speech and handwriting recognition fields. Yet due to the non-convexity of the optimization problem, previous works usually rely on severe approximations so that it is still an open problem. We propose a new learning algorithm that relies on non-convex optimization and bundle methods and allows tackling the original optimization problem as is. It is proved to converge to a solution with accuracy ɛ with a rate O(1/ɛ). We provide experimental results gained on speech and handwriting recognition that demonstrate the potential of the method. 1.
Keyphrases
partially observed state trinh-minh-tri hidden markov model large margin training open problem original optimization problem new learning algorithm optimization problem severe approximation previous work bundle method large margin learning recognition field experimental result non-convex optimization continuous density hmms
