Representing hypoexponential distributions in continuous time Bayesian networks

Manxia Liu; Fabio Stella; Arjen Hommersom; Peter J.F. Lucas

doi:10.1007/978-3-319-91479-4_47

Representing hypoexponential distributions in continuous time Bayesian networks

Manxia Liu^*, Fabio Stella, Arjen Hommersom, Peter J.F. Lucas

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)

Abstract

Continuous time Bayesian networks offer a compact representation for modeling structured stochastic processes that evolve over continuous time. In these models, the time duration that a variable stays in a state until a transition occurs is assumed to be exponentially distributed. In real-world scenarios, however, this assumption is rarely satisfied, in particular when describing more complex temporal processes. To relax this assumption, we propose an extension to support the modeling of the transitioning time as a hypoexponential distribution by introducing an additional hidden variable. Using such an approach, we also allow CTBNs to obtain memory, which is lacking in standard CTBNs. The parameter estimation in the proposed models is transformed into a learning task in their equivalent Markovian models.

Original language	English
Title of host publication	Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications
Subtitle of host publication	17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III
Editors	Irina Perfilieva, Jesus Medina, Manuel Ojeda-Aciego, Ronald R. Yager, Jose Luis Verdegay, Bernadette Bouchon-Meunier
Place of Publication	Cham
Publisher	Springer
Pages	565-577
Number of pages	13
ISBN (Electronic)	978-3-319-91479-4
ISBN (Print)	978-3-319-91478-7
DOIs	https://doi.org/10.1007/978-3-319-91479-4_47
Publication status	Published - 2018
Externally published	Yes
Event	17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018 - Cadiz, Spain Duration: 11 Jun 2018 → 15 Jun 2018 Conference number: 17

Publication series

Name	Communications in Computer and Information Science
Publisher	Springer
Volume	855
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018
Abbreviated title	IPMU 2018
Country/Territory	Spain
City	Cadiz
Period	11/06/18 → 15/06/18

Keywords

Continuous time bayesian networks
Dynamic Bayesian networks
Hidden variable
Memory
Phase-type distribution
n/a OA procedure

Access to Document

10.1007/978-3-319-91479-4_47

Cite this

Liu, M., Stella, F., Hommersom, A., & Lucas, P. J. F. (2018). Representing hypoexponential distributions in continuous time Bayesian networks. In I. Perfilieva, J. Medina, M. Ojeda-Aciego, R. R. Yager, J. L. Verdegay, & B. Bouchon-Meunier (Eds.), Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III (pp. 565-577). (Communications in Computer and Information Science; Vol. 855). Springer. https://doi.org/10.1007/978-3-319-91479-4_47

Liu, Manxia ; Stella, Fabio ; Hommersom, Arjen et al. / Representing hypoexponential distributions in continuous time Bayesian networks. Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. editor / Irina Perfilieva ; Jesus Medina ; Manuel Ojeda-Aciego ; Ronald R. Yager ; Jose Luis Verdegay ; Bernadette Bouchon-Meunier. Cham : Springer, 2018. pp. 565-577 (Communications in Computer and Information Science).

@inproceedings{1ce6f618475f4d348a5da35e85701dc1,

title = "Representing hypoexponential distributions in continuous time Bayesian networks",

abstract = "Continuous time Bayesian networks offer a compact representation for modeling structured stochastic processes that evolve over continuous time. In these models, the time duration that a variable stays in a state until a transition occurs is assumed to be exponentially distributed. In real-world scenarios, however, this assumption is rarely satisfied, in particular when describing more complex temporal processes. To relax this assumption, we propose an extension to support the modeling of the transitioning time as a hypoexponential distribution by introducing an additional hidden variable. Using such an approach, we also allow CTBNs to obtain memory, which is lacking in standard CTBNs. The parameter estimation in the proposed models is transformed into a learning task in their equivalent Markovian models.",

keywords = "Continuous time bayesian networks, Dynamic Bayesian networks, Hidden variable, Memory, Phase-type distribution, n/a OA procedure",

author = "Manxia Liu and Fabio Stella and Arjen Hommersom and Lucas, {Peter J.F.}",

note = "Funding Information: ML is supported by China Scholarship Council, by a grant from project NanoS-TIMA [NORTE-01-0145-FEDER-000016], which was financed by the North Portugal Regional Operational Programme [NORTE 2020], under the PORTUGAL 2020 Partnership Agreement, and through the European Regional Development Fund [ERDF], and by Italian Association for Artificial Intelligence (AI*IA). Publisher Copyright: {\textcopyright} Springer International Publishing AG, part of Springer Nature 2018.; 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018, IPMU 2018 ; Conference date: 11-06-2018 Through 15-06-2018",

year = "2018",

doi = "10.1007/978-3-319-91479-4_47",

language = "English",

isbn = "978-3-319-91478-7",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "565--577",

editor = "Irina Perfilieva and Jesus Medina and Manuel Ojeda-Aciego and Yager, {Ronald R.} and Verdegay, {Jose Luis} and Bernadette Bouchon-Meunier",

booktitle = "Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications",

address = "Germany",

}

Liu, M, Stella, F, Hommersom, A & Lucas, PJF 2018, Representing hypoexponential distributions in continuous time Bayesian networks. in I Perfilieva, J Medina, M Ojeda-Aciego, RR Yager, JL Verdegay & B Bouchon-Meunier (eds), Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. Communications in Computer and Information Science, vol. 855, Springer, Cham, pp. 565-577, 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018, Cadiz, Spain, 11/06/18. https://doi.org/10.1007/978-3-319-91479-4_47

Representing hypoexponential distributions in continuous time Bayesian networks. / Liu, Manxia; Stella, Fabio; Hommersom, Arjen et al.
Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. ed. / Irina Perfilieva; Jesus Medina; Manuel Ojeda-Aciego; Ronald R. Yager; Jose Luis Verdegay; Bernadette Bouchon-Meunier. Cham: Springer, 2018. p. 565-577 (Communications in Computer and Information Science; Vol. 855).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Representing hypoexponential distributions in continuous time Bayesian networks

AU - Liu, Manxia

AU - Stella, Fabio

AU - Hommersom, Arjen

AU - Lucas, Peter J.F.

N1 - Conference code: 17

PY - 2018

Y1 - 2018

N2 - Continuous time Bayesian networks offer a compact representation for modeling structured stochastic processes that evolve over continuous time. In these models, the time duration that a variable stays in a state until a transition occurs is assumed to be exponentially distributed. In real-world scenarios, however, this assumption is rarely satisfied, in particular when describing more complex temporal processes. To relax this assumption, we propose an extension to support the modeling of the transitioning time as a hypoexponential distribution by introducing an additional hidden variable. Using such an approach, we also allow CTBNs to obtain memory, which is lacking in standard CTBNs. The parameter estimation in the proposed models is transformed into a learning task in their equivalent Markovian models.

AB - Continuous time Bayesian networks offer a compact representation for modeling structured stochastic processes that evolve over continuous time. In these models, the time duration that a variable stays in a state until a transition occurs is assumed to be exponentially distributed. In real-world scenarios, however, this assumption is rarely satisfied, in particular when describing more complex temporal processes. To relax this assumption, we propose an extension to support the modeling of the transitioning time as a hypoexponential distribution by introducing an additional hidden variable. Using such an approach, we also allow CTBNs to obtain memory, which is lacking in standard CTBNs. The parameter estimation in the proposed models is transformed into a learning task in their equivalent Markovian models.

KW - Continuous time bayesian networks

KW - Dynamic Bayesian networks

KW - Hidden variable

KW - Memory

KW - Phase-type distribution

KW - n/a OA procedure

UR - http://www.scopus.com/inward/record.url?scp=85048033128&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-91479-4_47

DO - 10.1007/978-3-319-91479-4_47

M3 - Conference contribution

SN - 978-3-319-91478-7

T3 - Communications in Computer and Information Science

SP - 565

EP - 577

BT - Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications

A2 - Perfilieva, Irina

A2 - Medina, Jesus

A2 - Ojeda-Aciego, Manuel

A2 - Yager, Ronald R.

A2 - Verdegay, Jose Luis

A2 - Bouchon-Meunier, Bernadette

PB - Springer

CY - Cham

T2 - 17th International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, IPMU 2018

Y2 - 11 June 2018 through 15 June 2018

ER -

Liu M, Stella F, Hommersom A, Lucas PJF. Representing hypoexponential distributions in continuous time Bayesian networks. In Perfilieva I, Medina J, Ojeda-Aciego M, Yager RR, Verdegay JL, Bouchon-Meunier B, editors, Information Processing and Management of Uncertainty in Knowledge-Based Systems. Applications: 17th International Conference, IPMU 2018, Cádiz, Spain, June 11-15, 2018, Proceedings, Part III. Cham: Springer. 2018. p. 565-577. (Communications in Computer and Information Science). doi: 10.1007/978-3-319-91479-4_47

Representing hypoexponential distributions in continuous time Bayesian networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this