Linear Time Inference in Hierarchical HMMs

DMCA

Linear Time Inference in Hierarchical HMMs (2001)

Cached

Download Links

by Kevin P. Murphy , Mark A. Paskin

Venue:	In Proceedings of Neural Information Processing Systems
Citations:	108 - 3 self

BibTeX

@INPROCEEDINGS{Murphy01lineartime,
    author = {Kevin P. Murphy and Mark A. Paskin},
    title = {Linear Time Inference in Hierarchical HMMs},
    booktitle = {In Proceedings of Neural Information Processing Systems},
    year = {2001}
}

OpenURL

Abstract

The hierarchical hidden Markov model (HHMM) is a generalization of the hidden Markov model (HMM) that models sequences with structure at many length/time scales [FST98]. Unfortunately, the original inference algorithm is rather complicated, and takes O(T ) time, where T is the length of the sequence, making it impractical for many domains. In this paper, we show how HHMMs are a special kind of dynamic Bayesian network (DBN), and thereby derive a much simpler inference algorithm, which only takes O(T ) time. Furthermore, by drawing the connection between HHMMs and DBNs, we enable the application of many standard approximation techniques to further speed up inference.

Keyphrases

hierarchical hmms linear time inference model sequence original inference algorithm special kind inference algorithm hidden markov model many length time scale dynamic bayesian network many standard approximation technique hierarchical hidden markov model many domain

DMCA