Parallel importance sampling in conditional linear Gaussian networks

Antonio Salmerón, Darío Ramos-López, Hanen Borchani, Ana Maria Martinez, Andres R. Masegosa, Antonio Fernández, Helge Langseth, Anders Læsø Madsen, Thomas Dyhre Nielsen

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

6 Citationer (Scopus)
269 Downloads (Pure)

Resumé

In this paper we analyse the problem of probabilistic inference in CLG networks when evidence comes in streams. In such situations, fast and scalable algorithms, able to provide accurate responses in a short time are required. We consider the instantiation of variational inference and importance sampling, two well known tools for probabilistic inference, to the CLG case. The experimental results over synthetic networks show how a parallel version importance sampling, and more precisely evidence weighting, is a promising scheme, as it is accurate and scales up with respect to available computing resources.
OriginalsprogEngelsk
TitelAdvances in Artificial Intelligence : 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings
RedaktørerJosé M. Puerta, José A. Gámez, Bernabe Dorronsoro, Edurne Barrenechea, Alicia Troncoso, Bruno Baruque, Mikel Galar
ForlagSpringer
Publikationsdato2015
Sider36-46
ISBN (Trykt)978-3-319-24597-3
ISBN (Elektronisk)978-3-319-24598-0
DOI
StatusUdgivet - 2015
BegivenhedConference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 - Albacete, Spanien
Varighed: 9 nov. 201512 nov. 2015
Konferencens nummer: 16th

Konference

KonferenceConference of the Spanish Association for Artificial Intelligence, CAEPIA 2015
Nummer16th
LandSpanien
ByAlbacete
Periode09/11/201512/11/2015
NavnLecture Notes in Computer Science
Nummer9422
ISSN0302-9743

Fingerprint

Importance sampling

Citer dette

Salmerón, A., Ramos-López, D., Borchani, H., Martinez, A. M., Masegosa, A. R., Fernández, A., ... Nielsen, T. D. (2015). Parallel importance sampling in conditional linear Gaussian networks. I J. M. Puerta, J. A. Gámez, B. Dorronsoro, E. Barrenechea, A. Troncoso, B. Baruque, & M. Galar (red.), Advances in Artificial Intelligence: 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings (s. 36-46). Springer. Lecture Notes in Computer Science, Nr. 9422 https://doi.org/10.1007/978-3-319-24598-0_4
Salmerón, Antonio ; Ramos-López, Darío ; Borchani, Hanen ; Martinez, Ana Maria ; Masegosa, Andres R. ; Fernández, Antonio ; Langseth, Helge ; Madsen, Anders Læsø ; Nielsen, Thomas Dyhre. / Parallel importance sampling in conditional linear Gaussian networks. Advances in Artificial Intelligence: 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings. red. / José M. Puerta ; José A. Gámez ; Bernabe Dorronsoro ; Edurne Barrenechea ; Alicia Troncoso ; Bruno Baruque ; Mikel Galar. Springer, 2015. s. 36-46 (Lecture Notes in Computer Science; Nr. 9422).
@inproceedings{8438e1470d26401590b173dd32942fb4,
title = "Parallel importance sampling in conditional linear Gaussian networks",
abstract = "In this paper we analyse the problem of probabilistic inference in CLG networks when evidence comes in streams. In such situations, fast and scalable algorithms, able to provide accurate responses in a short time are required. We consider the instantiation of variational inference and importance sampling, two well known tools for probabilistic inference, to the CLG case. The experimental results over synthetic networks show how a parallel version importance sampling, and more precisely evidence weighting, is a promising scheme, as it is accurate and scales up with respect to available computing resources.",
author = "Antonio Salmer{\'o}n and Dar{\'i}o Ramos-L{\'o}pez and Hanen Borchani and Martinez, {Ana Maria} and Masegosa, {Andres R.} and Antonio Fern{\'a}ndez and Helge Langseth and Madsen, {Anders L{\ae}s{\o}} and Nielsen, {Thomas Dyhre}",
year = "2015",
doi = "10.1007/978-3-319-24598-0_4",
language = "English",
isbn = "978-3-319-24597-3",
pages = "36--46",
editor = "Puerta, {Jos{\'e} M.} and G{\'a}mez, {Jos{\'e} A.} and Bernabe Dorronsoro and Edurne Barrenechea and Alicia Troncoso and Bruno Baruque and Mikel Galar",
booktitle = "Advances in Artificial Intelligence",
publisher = "Springer",
address = "Germany",

}

Salmerón, A, Ramos-López, D, Borchani, H, Martinez, AM, Masegosa, AR, Fernández, A, Langseth, H, Madsen, AL & Nielsen, TD 2015, Parallel importance sampling in conditional linear Gaussian networks. i JM Puerta, JA Gámez, B Dorronsoro, E Barrenechea, A Troncoso, B Baruque & M Galar (red), Advances in Artificial Intelligence: 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings. Springer, Lecture Notes in Computer Science, nr. 9422, s. 36-46, Albacete, Spanien, 09/11/2015. https://doi.org/10.1007/978-3-319-24598-0_4

Parallel importance sampling in conditional linear Gaussian networks. / Salmerón, Antonio; Ramos-López, Darío ; Borchani, Hanen; Martinez, Ana Maria; Masegosa, Andres R.; Fernández, Antonio; Langseth, Helge; Madsen, Anders Læsø; Nielsen, Thomas Dyhre.

Advances in Artificial Intelligence: 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings. red. / José M. Puerta; José A. Gámez; Bernabe Dorronsoro; Edurne Barrenechea; Alicia Troncoso; Bruno Baruque; Mikel Galar. Springer, 2015. s. 36-46 (Lecture Notes in Computer Science; Nr. 9422).

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Parallel importance sampling in conditional linear Gaussian networks

AU - Salmerón, Antonio

AU - Ramos-López, Darío

AU - Borchani, Hanen

AU - Martinez, Ana Maria

AU - Masegosa, Andres R.

AU - Fernández, Antonio

AU - Langseth, Helge

AU - Madsen, Anders Læsø

AU - Nielsen, Thomas Dyhre

PY - 2015

Y1 - 2015

N2 - In this paper we analyse the problem of probabilistic inference in CLG networks when evidence comes in streams. In such situations, fast and scalable algorithms, able to provide accurate responses in a short time are required. We consider the instantiation of variational inference and importance sampling, two well known tools for probabilistic inference, to the CLG case. The experimental results over synthetic networks show how a parallel version importance sampling, and more precisely evidence weighting, is a promising scheme, as it is accurate and scales up with respect to available computing resources.

AB - In this paper we analyse the problem of probabilistic inference in CLG networks when evidence comes in streams. In such situations, fast and scalable algorithms, able to provide accurate responses in a short time are required. We consider the instantiation of variational inference and importance sampling, two well known tools for probabilistic inference, to the CLG case. The experimental results over synthetic networks show how a parallel version importance sampling, and more precisely evidence weighting, is a promising scheme, as it is accurate and scales up with respect to available computing resources.

U2 - 10.1007/978-3-319-24598-0_4

DO - 10.1007/978-3-319-24598-0_4

M3 - Article in proceeding

SN - 978-3-319-24597-3

SP - 36

EP - 46

BT - Advances in Artificial Intelligence

A2 - Puerta, José M.

A2 - Gámez, José A.

A2 - Dorronsoro, Bernabe

A2 - Barrenechea, Edurne

A2 - Troncoso, Alicia

A2 - Baruque, Bruno

A2 - Galar, Mikel

PB - Springer

ER -

Salmerón A, Ramos-López D, Borchani H, Martinez AM, Masegosa AR, Fernández A et al. Parallel importance sampling in conditional linear Gaussian networks. I Puerta JM, Gámez JA, Dorronsoro B, Barrenechea E, Troncoso A, Baruque B, Galar M, red., Advances in Artificial Intelligence: 16th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2015 Albacete, Spain, November 9–12, 2015 Proceedings. Springer. 2015. s. 36-46. (Lecture Notes in Computer Science; Nr. 9422). https://doi.org/10.1007/978-3-319-24598-0_4