Multi-classifiers of Small Treewidth

A.J. Pastink; L.C. van der Gaag

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

Multi-classifiers of Small Treewidth

A.J. Pastink
, L.C. van der Gaag

Dept of Information and Computing Sciences

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

Abstract

Multi-dimensional Bayesian network classifiers are becoming quite popular for multi-label classification. These models have the advantage of a high expressive power, but may induce a prohibitively high runtime of classification. We argue that the high runtime burden originates from their large treewidth. Thus motivated, we present an algorithm for learning multi-classifiers of small treewidth. Experimental results show that these models have a small runtime of classification, without loosing accuracy compared to unconstrained multi-classifiers.

Original language	English
Title of host publication	ECSQARU 2015: 13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty
Editors	Sebastien Destercke, Thierry Denoeux
Publisher	Springer
Pages	199-209
Number of pages	11
Volume	9161
ISBN (Electronic)	9783319208077
ISBN (Print)	9783319208060
DOIs	https://doi.org/10.1007/978-3-319-20807-7
Publication status	Published - 2015

Access to Document

10.1007/978-3-319-20807-7

treewidthFinal published version, 272 KB

Cite this

@inproceedings{4b863f2c9b8d4960a8bbc50ef1d2d405,

title = "Multi-classifiers of Small Treewidth",

abstract = "Multi-dimensional Bayesian network classifiers are becoming quite popular for multi-label classification. These models have the advantage of a high expressive power, but may induce a prohibitively high runtime of classification. We argue that the high runtime burden originates from their large treewidth. Thus motivated, we present an algorithm for learning multi-classifiers of small treewidth. Experimental results show that these models have a small runtime of classification, without loosing accuracy compared to unconstrained multi-classifiers.",

author = "A.J. Pastink and \{van der Gaag\}, L.C.",

year = "2015",

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

language = "English",

isbn = "9783319208060",

volume = "9161",

pages = "199--209",

editor = "Sebastien Destercke and Thierry Denoeux",

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

publisher = "Springer",

}

Multi-classifiers of Small Treewidth. / Pastink, A.J.; van der Gaag, L.C.
ECSQARU 2015: 13th European Conference on Symbolic and Quantitative Approaches to Reasoning with Uncertainty. ed. / Sebastien Destercke; Thierry Denoeux. Vol. 9161 Springer, 2015. p. 199-209.

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

TY - GEN

T1 - Multi-classifiers of Small Treewidth

AU - Pastink, A.J.

AU - van der Gaag, L.C.

PY - 2015

Y1 - 2015

N2 - Multi-dimensional Bayesian network classifiers are becoming quite popular for multi-label classification. These models have the advantage of a high expressive power, but may induce a prohibitively high runtime of classification. We argue that the high runtime burden originates from their large treewidth. Thus motivated, we present an algorithm for learning multi-classifiers of small treewidth. Experimental results show that these models have a small runtime of classification, without loosing accuracy compared to unconstrained multi-classifiers.

AB - Multi-dimensional Bayesian network classifiers are becoming quite popular for multi-label classification. These models have the advantage of a high expressive power, but may induce a prohibitively high runtime of classification. We argue that the high runtime burden originates from their large treewidth. Thus motivated, we present an algorithm for learning multi-classifiers of small treewidth. Experimental results show that these models have a small runtime of classification, without loosing accuracy compared to unconstrained multi-classifiers.

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

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

M3 - Conference contribution

SN - 9783319208060

VL - 9161

SP - 199

EP - 209

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

A2 - Destercke, Sebastien

A2 - Denoeux, Thierry

PB - Springer

ER -

Multi-classifiers of Small Treewidth

Abstract

Access to Document

Fingerprint

Cite this