Entropy computation in partially observed Markov chains

François Desbouvries

doi:10.1063/1.2423294

Entropy computation in partially observed Markov chains

François Desbouvries

CNRS SAMOVAR UMR 5157

Research output: Contribution to journal › Article › peer-review

Abstract

Let X = X_nn∈N be a hidden process and Y = Y _nn∈N be an observed process. We assume that (X,Y) is a (pairwise) Markov Chain (PMC). PMC are more general than Hidden Markov Chains (HMC) and yet enable the development of efficient parameter estimation and Bayesian restoration algorithms. In this paper we propose a fast (i.e., O(N)) algorithm for computing the entropy of X_nn=0^N given an observation sequence y_nn=0^N.

Original language	English
Pages (from-to)	355-357
Number of pages	3
Journal	AIP Conference Proceedings
Volume	872
DOIs	https://doi.org/10.1063/1.2423294
Publication status	Published - 27 Dec 2006
Externally published	Yes

Keywords

Entropy
Hidden Markov models
Partially observed Markov chains

Access to Document

10.1063/1.2423294

Cite this

@article{1327c37177c644eda6b6a1f20dc0c433,

title = "Entropy computation in partially observed Markov chains",

abstract = "Let X = Xnn∈N be a hidden process and Y = Y nn∈N be an observed process. We assume that (X,Y) is a (pairwise) Markov Chain (PMC). PMC are more general than Hidden Markov Chains (HMC) and yet enable the development of efficient parameter estimation and Bayesian restoration algorithms. In this paper we propose a fast (i.e., O(N)) algorithm for computing the entropy of Xnn=0N given an observation sequence ynn=0N.",

keywords = "Entropy, Hidden Markov models, Partially observed Markov chains",

author = "Fran{\c c}ois Desbouvries",

year = "2006",

month = dec,

day = "27",

doi = "10.1063/1.2423294",

language = "English",

volume = "872",

pages = "355--357",

journal = "AIP Conference Proceedings",

issn = "0094-243X",

}

TY - JOUR

T1 - Entropy computation in partially observed Markov chains

AU - Desbouvries, François

PY - 2006/12/27

Y1 - 2006/12/27

N2 - Let X = Xnn∈N be a hidden process and Y = Y nn∈N be an observed process. We assume that (X,Y) is a (pairwise) Markov Chain (PMC). PMC are more general than Hidden Markov Chains (HMC) and yet enable the development of efficient parameter estimation and Bayesian restoration algorithms. In this paper we propose a fast (i.e., O(N)) algorithm for computing the entropy of Xnn=0N given an observation sequence ynn=0N.

AB - Let X = Xnn∈N be a hidden process and Y = Y nn∈N be an observed process. We assume that (X,Y) is a (pairwise) Markov Chain (PMC). PMC are more general than Hidden Markov Chains (HMC) and yet enable the development of efficient parameter estimation and Bayesian restoration algorithms. In this paper we propose a fast (i.e., O(N)) algorithm for computing the entropy of Xnn=0N given an observation sequence ynn=0N.

KW - Entropy

KW - Hidden Markov models

KW - Partially observed Markov chains

U2 - 10.1063/1.2423294

DO - 10.1063/1.2423294

M3 - Article

AN - SCOPUS:33845670494

SN - 0094-243X

VL - 872

SP - 355

EP - 357

JO - AIP Conference Proceedings

JF - AIP Conference Proceedings

ER -

Entropy computation in partially observed Markov chains

Abstract

Keywords

Access to Document

Fingerprint

Cite this