MAP segmentation in Bayesian hidden Markov models : a case study. / Koloydenko, Alexey; Kuljus, Kristi; Lember, Jüri.

In: Journal of Applied Statistics, 10.12.2020, p. 1-32.

Research output: Contribution to journalArticlepeer-review

E-pub ahead of print

Standard

MAP segmentation in Bayesian hidden Markov models : a case study. / Koloydenko, Alexey; Kuljus, Kristi; Lember, Jüri.

In: Journal of Applied Statistics, 10.12.2020, p. 1-32.

Research output: Contribution to journalArticlepeer-review

Harvard

APA

Vancouver

Author

Koloydenko, Alexey ; Kuljus, Kristi ; Lember, Jüri. / MAP segmentation in Bayesian hidden Markov models : a case study. In: Journal of Applied Statistics. 2020 ; pp. 1-32.

BibTeX

@article{9ce6ea5ff1b24e16954a6152b1da194b,
title = "MAP segmentation in Bayesian hidden Markov models: a case study",
abstract = "We consider the problem of estimating the maximum posterior probability (MAP)state sequence for a finite state and finite emission alphabet hidden Markov model(HMM) in the Bayesian setup, where both emission and transition matrices haveDirichlet priors. We study a training set consisting of thousands of protein alignment pairs. The training data is used to set the prior hyperparameters for Bayesian MAP segmentation. Since the Viterbi algorithm is not applicable any more, there is no simple procedure to find the MAP path, and several iterative algorithms are considered and compared. The main goal of the paper is to test the Bayesian setup against the frequentist one, where the parameters of HMM are estimated using the training data.",
author = "Alexey Koloydenko and Kristi Kuljus and J{\"u}ri Lember",
year = "2020",
month = dec,
day = "10",
doi = "10.1080/02664763.2020.1858273",
language = "English",
pages = "1--32",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",

}

RIS

TY - JOUR

T1 - MAP segmentation in Bayesian hidden Markov models

T2 - a case study

AU - Koloydenko, Alexey

AU - Kuljus, Kristi

AU - Lember, Jüri

PY - 2020/12/10

Y1 - 2020/12/10

N2 - We consider the problem of estimating the maximum posterior probability (MAP)state sequence for a finite state and finite emission alphabet hidden Markov model(HMM) in the Bayesian setup, where both emission and transition matrices haveDirichlet priors. We study a training set consisting of thousands of protein alignment pairs. The training data is used to set the prior hyperparameters for Bayesian MAP segmentation. Since the Viterbi algorithm is not applicable any more, there is no simple procedure to find the MAP path, and several iterative algorithms are considered and compared. The main goal of the paper is to test the Bayesian setup against the frequentist one, where the parameters of HMM are estimated using the training data.

AB - We consider the problem of estimating the maximum posterior probability (MAP)state sequence for a finite state and finite emission alphabet hidden Markov model(HMM) in the Bayesian setup, where both emission and transition matrices haveDirichlet priors. We study a training set consisting of thousands of protein alignment pairs. The training data is used to set the prior hyperparameters for Bayesian MAP segmentation. Since the Viterbi algorithm is not applicable any more, there is no simple procedure to find the MAP path, and several iterative algorithms are considered and compared. The main goal of the paper is to test the Bayesian setup against the frequentist one, where the parameters of HMM are estimated using the training data.

UR - https://arxiv.org/abs/2004.08336

U2 - 10.1080/02664763.2020.1858273

DO - 10.1080/02664763.2020.1858273

M3 - Article

SP - 1

EP - 32

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

ER -