ML Anthology

Linear-Time Inference in Hierarchical HMMs

Kevin P. Murphy, Mark A. Paskin
NeurIPS 2001 pp. 833-840
/neurips/2001/murphy2001neurips-lineartime/

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 infer- is ence algorithm is rather complicated, and takes 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 time. Furthermore, by drawing the connection between HHMMs and DBNs, we enable the application of many stan- dard approximation techniques to further speed up inference.

PDF NeurIPS Semantic Scholar

Cite

Text

Murphy and Paskin. "Linear-Time Inference in Hierarchical HMMs." Neural Information Processing Systems, 2001.

Markdown

[Murphy and Paskin. "Linear-Time Inference in Hierarchical HMMs." Neural Information Processing Systems, 2001.](https://mlanthology.org/neurips/2001/murphy2001neurips-lineartime/)

BibTeX

@inproceedings{murphy2001neurips-lineartime,
  title     = {{Linear-Time Inference in Hierarchical HMMs}},
  author    = {Murphy, Kevin P. and Paskin, Mark A.},
  booktitle = {Neural Information Processing Systems},
  year      = {2001},
  pages     = {833-840},
  url       = {https://mlanthology.org/neurips/2001/murphy2001neurips-lineartime/}
}