Identifying combinatorially symmetric Hidden Markov Models

Daniel Burgarth

doi:10.13001/1081-3810.3651

Identifying combinatorially symmetric Hidden Markov Models

Daniel Burgarth

Mathemateg

Allbwn ymchwil: Cyfraniad at gyfnodolyn › Erthygl › adolygiad gan gymheiriaid

Crynodeb

A sufficient criterion for the unique parameter identification of combinatorially symmetric Hidden Markov Models, based on the structure of their transition matrix, is provided. If the observed states of the chain form a zero forcing set of the graph of the Markov model, then it is uniquely identifiable and an explicit reconstruction method is given.

Iaith wreiddiol	Saesneg
Tudalennau (o-i)	393-398
Cyfnodolyn	Electronic Journal of Linear Algebra
Cyfrol	34
Rhif cyhoeddi	1
Dynodwyr Gwrthrych Digidol (DOIs)	https://doi.org/10.13001/1081-3810.3651
Statws	Cyhoeddwyd - Awst 2018

Mynediad at Ddogfen

10.13001/1081-3810.3651

https://repository.uwyo.edu/ela/vol34/iss1/30

Cysylltiad parhaol

Dolen barhaus

Dyfynnu hyn

@article{3cc0d657ae024082ba4ddb69af35bf1e,

title = "Identifying combinatorially symmetric Hidden Markov Models",

abstract = "A sufficient criterion for the unique parameter identification of combinatorially symmetric Hidden Markov Models, based on the structure of their transition matrix, is provided. If the observed states of the chain form a zero forcing set of the graph of the Markov model, then it is uniquely identifiable and an explicit reconstruction method is given.",

keywords = "Markov chains, zero facing, parameter identification",

author = "Daniel Burgarth",

year = "2018",

month = aug,

doi = "10.13001/1081-3810.3651",

language = "English",

volume = "34",

pages = "393--398",

journal = "Electronic Journal of Linear Algebra",

issn = "1081-3810",

publisher = "International Linear Algebra Society",

number = "1",

}

TY - JOUR

T1 - Identifying combinatorially symmetric Hidden Markov Models

AU - Burgarth, Daniel

PY - 2018/8

Y1 - 2018/8

N2 - A sufficient criterion for the unique parameter identification of combinatorially symmetric Hidden Markov Models, based on the structure of their transition matrix, is provided. If the observed states of the chain form a zero forcing set of the graph of the Markov model, then it is uniquely identifiable and an explicit reconstruction method is given.

AB - A sufficient criterion for the unique parameter identification of combinatorially symmetric Hidden Markov Models, based on the structure of their transition matrix, is provided. If the observed states of the chain form a zero forcing set of the graph of the Markov model, then it is uniquely identifiable and an explicit reconstruction method is given.

KW - Markov chains

KW - zero facing

KW - parameter identification

U2 - 10.13001/1081-3810.3651

DO - 10.13001/1081-3810.3651

M3 - Article

SN - 1081-3810

VL - 34

SP - 393

EP - 398

JO - Electronic Journal of Linear Algebra

JF - Electronic Journal of Linear Algebra

IS - 1

ER -

Identifying combinatorially symmetric Hidden Markov Models

Crynodeb

Mynediad at Ddogfen

Cysylltiad parhaol

Ôl bys

Dyfynnu hyn