Hybrid time bayesian networks

Manxia Liu; Arjen Hommersom; Maarten van der Heijden; Peter J.F. Lucas

doi:10.1007/978-3-319-20807-7_34

Hybrid time bayesian networks

Manxia Liu^*, Arjen Hommersom, Maarten van der Heijden, Peter J.F. Lucas

^*Corresponding author for this work

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

3 Citations (Scopus)

Abstract

Capturing heterogeneous dynamic systems in a probabilistic model is a challenging problem. A single time granularity, such as employed by dynamic Bayesian networks, provides insufficient flexibility to capture the dynamics of many real-world processes. The alternative is to assume that time is continuous, giving rise to continuous time Bayesian networks. Here the problem is that the level of temporal detail is too precise to match available probabilistic knowledge. In this paper, we present a novel class of models, called hybrid time Bayesian networks, which combine discrete-time and continuous-time Bayesian networks. The new formalism allows us to more naturally model dynamic systems with regular and irregularly changing variables. Its usefulness is illustrated by means of a real-world medical problem.

Original language	English
Title of host publication	Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings
Editors	Sébastien Destercke, Thierry Denoeux
Publisher	Springer
Pages	376-386
Number of pages	11
ISBN (Print)	9783319208060
DOIs	https://doi.org/10.1007/978-3-319-20807-7_34
Publication status	Published - 2015
Externally published	Yes
Event	13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU 2015 - Compiègne, France Duration: 15 Jul 2015 → 17 Jul 2015 Conference number: 13

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9161
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU 2015
Abbreviated title	ECSQARU 2015
Country/Territory	France
City	Compiègne
Period	15/07/15 → 17/07/15

Keywords

n/a OA procedure
Dynamic Bayesian networks
Dynamic systems
Continuous time Bayesian networks

Access to Document

10.1007/978-3-319-20807-7_34

Cite this

Liu, M., Hommersom, A., van der Heijden, M., & Lucas, P. J. F. (2015). Hybrid time bayesian networks. In S. Destercke, & T. Denoeux (Eds.), Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings (pp. 376-386). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9161). Springer. https://doi.org/10.1007/978-3-319-20807-7_34

Liu, Manxia ; Hommersom, Arjen ; van der Heijden, Maarten et al. / Hybrid time bayesian networks. Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings. editor / Sébastien Destercke ; Thierry Denoeux. Springer, 2015. pp. 376-386 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{efb23c7fa4634c3f9c370aa169f8fa9d,

title = "Hybrid time bayesian networks",

abstract = "Capturing heterogeneous dynamic systems in a probabilistic model is a challenging problem. A single time granularity, such as employed by dynamic Bayesian networks, provides insufficient flexibility to capture the dynamics of many real-world processes. The alternative is to assume that time is continuous, giving rise to continuous time Bayesian networks. Here the problem is that the level of temporal detail is too precise to match available probabilistic knowledge. In this paper, we present a novel class of models, called hybrid time Bayesian networks, which combine discrete-time and continuous-time Bayesian networks. The new formalism allows us to more naturally model dynamic systems with regular and irregularly changing variables. Its usefulness is illustrated by means of a real-world medical problem.",

keywords = "n/a OA procedure, Dynamic Bayesian networks, Dynamic systems, Continuous time Bayesian networks",

author = "Manxia Liu and Arjen Hommersom and {van der Heijden}, Maarten and Lucas, {Peter J.F.}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU 2015, ECSQARU 2015 ; Conference date: 15-07-2015 Through 17-07-2015",

year = "2015",

doi = "10.1007/978-3-319-20807-7_34",

language = "English",

isbn = "9783319208060",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "376--386",

editor = "S{\'e}bastien Destercke and Thierry Denoeux",

booktitle = "Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings",

address = "Germany",

}

Liu, M, Hommersom, A, van der Heijden, M & Lucas, PJF 2015, Hybrid time bayesian networks. in S Destercke & T Denoeux (eds), Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9161, Springer, pp. 376-386, 13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU 2015, Compiègne, France, 15/07/15. https://doi.org/10.1007/978-3-319-20807-7_34

Hybrid time bayesian networks. / Liu, Manxia; Hommersom, Arjen; van der Heijden, Maarten et al.
Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings. ed. / Sébastien Destercke; Thierry Denoeux. Springer, 2015. p. 376-386 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9161).

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

TY - GEN

T1 - Hybrid time bayesian networks

AU - Liu, Manxia

AU - Hommersom, Arjen

AU - van der Heijden, Maarten

AU - Lucas, Peter J.F.

N1 - Conference code: 13

PY - 2015

Y1 - 2015

N2 - Capturing heterogeneous dynamic systems in a probabilistic model is a challenging problem. A single time granularity, such as employed by dynamic Bayesian networks, provides insufficient flexibility to capture the dynamics of many real-world processes. The alternative is to assume that time is continuous, giving rise to continuous time Bayesian networks. Here the problem is that the level of temporal detail is too precise to match available probabilistic knowledge. In this paper, we present a novel class of models, called hybrid time Bayesian networks, which combine discrete-time and continuous-time Bayesian networks. The new formalism allows us to more naturally model dynamic systems with regular and irregularly changing variables. Its usefulness is illustrated by means of a real-world medical problem.

AB - Capturing heterogeneous dynamic systems in a probabilistic model is a challenging problem. A single time granularity, such as employed by dynamic Bayesian networks, provides insufficient flexibility to capture the dynamics of many real-world processes. The alternative is to assume that time is continuous, giving rise to continuous time Bayesian networks. Here the problem is that the level of temporal detail is too precise to match available probabilistic knowledge. In this paper, we present a novel class of models, called hybrid time Bayesian networks, which combine discrete-time and continuous-time Bayesian networks. The new formalism allows us to more naturally model dynamic systems with regular and irregularly changing variables. Its usefulness is illustrated by means of a real-world medical problem.

KW - n/a OA procedure

KW - Dynamic Bayesian networks

KW - Dynamic systems

KW - Continuous time Bayesian networks

UR - http://www.scopus.com/inward/record.url?scp=84950321924&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-20807-7_34

DO - 10.1007/978-3-319-20807-7_34

M3 - Conference contribution

AN - SCOPUS:84950321924

SN - 9783319208060

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 376

EP - 386

BT - Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings

A2 - Destercke, Sébastien

A2 - Denoeux, Thierry

PB - Springer

T2 - 13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty, ECSQARU 2015

Y2 - 15 July 2015 through 17 July 2015

ER -

Liu M, Hommersom A, van der Heijden M, Lucas PJF. Hybrid time bayesian networks. In Destercke S, Denoeux T, editors, Symbolic and Quantitative Approaches to Reasoning with Uncertainty - 13th European Conference, ECSQARU 2015, Proceedings. Springer. 2015. p. 376-386. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-20807-7_34

Hybrid time bayesian networks

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this