AUC Maximization in Bayesian hierarchical models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The area under the curve (AUC) measures such as the area under the receiver operating characteristics curve (AUROC) and the area under the precision-recall curve (AUPR) are known to be more appropriate than the error rate, especially, for imbalanced data sets. There are several algorithms to optimize AUC measures instead of minimizing the error rate. However, this idea has not been fully exploited in Bayesian hierarchical models owing to the difficulties in inference. Here, we formulate a general Bayesian inference framework, called Bayesian AUC Maximization (BAM), to integrate AUC maximization into Bayesian hierarchical models by borrowing the pairwise and listwise ranking ideas from the information retrieval literature. To showcase our BAM framework, we develop two Bayesian linear classifier variants for two ranking approaches and derive their variational inference procedures. We perform validation experiments on four biomedical data sets to demonstrate the better predictive performance of our framework over its error-minimizing counterpart in terms of average AUROC and AUPR values.

Original languageEnglish (US)
Title of host publicationFrontiers in Artificial Intelligence and Applications
PublisherIOS Press
Pages21-27
Number of pages7
Volume285
ISBN (Electronic)9781614996712
DOIs
StatePublished - 2016
Externally publishedYes
Event22nd European Conference on Artificial Intelligence, ECAI 2016 - The Hague, Netherlands
Duration: Aug 29 2016Sep 2 2016

Publication series

NameFrontiers in Artificial Intelligence and Applications
Volume285
ISSN (Print)09226389

Other

Other22nd European Conference on Artificial Intelligence, ECAI 2016
CountryNetherlands
CityThe Hague
Period8/29/169/2/16

Fingerprint

Information retrieval
Classifiers
Experiments

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Gonen, M. (2016). AUC Maximization in Bayesian hierarchical models. In Frontiers in Artificial Intelligence and Applications (Vol. 285, pp. 21-27). (Frontiers in Artificial Intelligence and Applications; Vol. 285). IOS Press. https://doi.org/10.3233/978-1-61499-672-9-21

AUC Maximization in Bayesian hierarchical models. / Gonen, Mehmet.

Frontiers in Artificial Intelligence and Applications. Vol. 285 IOS Press, 2016. p. 21-27 (Frontiers in Artificial Intelligence and Applications; Vol. 285).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Gonen, M 2016, AUC Maximization in Bayesian hierarchical models. in Frontiers in Artificial Intelligence and Applications. vol. 285, Frontiers in Artificial Intelligence and Applications, vol. 285, IOS Press, pp. 21-27, 22nd European Conference on Artificial Intelligence, ECAI 2016, The Hague, Netherlands, 8/29/16. https://doi.org/10.3233/978-1-61499-672-9-21
Gonen M. AUC Maximization in Bayesian hierarchical models. In Frontiers in Artificial Intelligence and Applications. Vol. 285. IOS Press. 2016. p. 21-27. (Frontiers in Artificial Intelligence and Applications). https://doi.org/10.3233/978-1-61499-672-9-21
Gonen, Mehmet. / AUC Maximization in Bayesian hierarchical models. Frontiers in Artificial Intelligence and Applications. Vol. 285 IOS Press, 2016. pp. 21-27 (Frontiers in Artificial Intelligence and Applications).
@inproceedings{34071957a2044db4802af5798246d58c,
title = "AUC Maximization in Bayesian hierarchical models",
abstract = "The area under the curve (AUC) measures such as the area under the receiver operating characteristics curve (AUROC) and the area under the precision-recall curve (AUPR) are known to be more appropriate than the error rate, especially, for imbalanced data sets. There are several algorithms to optimize AUC measures instead of minimizing the error rate. However, this idea has not been fully exploited in Bayesian hierarchical models owing to the difficulties in inference. Here, we formulate a general Bayesian inference framework, called Bayesian AUC Maximization (BAM), to integrate AUC maximization into Bayesian hierarchical models by borrowing the pairwise and listwise ranking ideas from the information retrieval literature. To showcase our BAM framework, we develop two Bayesian linear classifier variants for two ranking approaches and derive their variational inference procedures. We perform validation experiments on four biomedical data sets to demonstrate the better predictive performance of our framework over its error-minimizing counterpart in terms of average AUROC and AUPR values.",
author = "Mehmet Gonen",
year = "2016",
doi = "10.3233/978-1-61499-672-9-21",
language = "English (US)",
volume = "285",
series = "Frontiers in Artificial Intelligence and Applications",
publisher = "IOS Press",
pages = "21--27",
booktitle = "Frontiers in Artificial Intelligence and Applications",

}

TY - GEN

T1 - AUC Maximization in Bayesian hierarchical models

AU - Gonen, Mehmet

PY - 2016

Y1 - 2016

N2 - The area under the curve (AUC) measures such as the area under the receiver operating characteristics curve (AUROC) and the area under the precision-recall curve (AUPR) are known to be more appropriate than the error rate, especially, for imbalanced data sets. There are several algorithms to optimize AUC measures instead of minimizing the error rate. However, this idea has not been fully exploited in Bayesian hierarchical models owing to the difficulties in inference. Here, we formulate a general Bayesian inference framework, called Bayesian AUC Maximization (BAM), to integrate AUC maximization into Bayesian hierarchical models by borrowing the pairwise and listwise ranking ideas from the information retrieval literature. To showcase our BAM framework, we develop two Bayesian linear classifier variants for two ranking approaches and derive their variational inference procedures. We perform validation experiments on four biomedical data sets to demonstrate the better predictive performance of our framework over its error-minimizing counterpart in terms of average AUROC and AUPR values.

AB - The area under the curve (AUC) measures such as the area under the receiver operating characteristics curve (AUROC) and the area under the precision-recall curve (AUPR) are known to be more appropriate than the error rate, especially, for imbalanced data sets. There are several algorithms to optimize AUC measures instead of minimizing the error rate. However, this idea has not been fully exploited in Bayesian hierarchical models owing to the difficulties in inference. Here, we formulate a general Bayesian inference framework, called Bayesian AUC Maximization (BAM), to integrate AUC maximization into Bayesian hierarchical models by borrowing the pairwise and listwise ranking ideas from the information retrieval literature. To showcase our BAM framework, we develop two Bayesian linear classifier variants for two ranking approaches and derive their variational inference procedures. We perform validation experiments on four biomedical data sets to demonstrate the better predictive performance of our framework over its error-minimizing counterpart in terms of average AUROC and AUPR values.

UR - http://www.scopus.com/inward/record.url?scp=85013042405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85013042405&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-672-9-21

DO - 10.3233/978-1-61499-672-9-21

M3 - Conference contribution

AN - SCOPUS:85013042405

VL - 285

T3 - Frontiers in Artificial Intelligence and Applications

SP - 21

EP - 27

BT - Frontiers in Artificial Intelligence and Applications

PB - IOS Press

ER -