Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach

Doris Xin, Ahmed El-Kishky, De Liao, Brandon Norick, Jiawei Han

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

Abstract

Active learning exploits inherent structures in the unlabeled data to minimize the number of labels required to train an accurate model. It enables effective machine learning in applications with high labeling cost, such as document classification and drug response prediction. We investigate active learning on heterogeneous information networks, with the objective of obtaining accurate node classifications while minimizing the number of labeled nodes. Our proposed algorithm harnesses a multi-armed bandit (MAB) algorithm to determine network structures that identify the most important nodes to the classification task, accounting for node types and without assuming label assortativity. Evaluations on real-world network classification tasks demonstrate that our algorithm outperforms existing methods independent of the underlying classification model.

Original languageEnglish (US)
Title of host publication2018 IEEE International Conference on Data Mining, ICDM 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1350-1355
Number of pages6
ISBN (Electronic)9781538691588
DOIs
StatePublished - Dec 27 2018
Event18th IEEE International Conference on Data Mining, ICDM 2018 - Singapore, Singapore
Duration: Nov 17 2018Nov 20 2018

Publication series

NameProceedings - IEEE International Conference on Data Mining, ICDM
Volume2018-November
ISSN (Print)1550-4786

Conference

Conference18th IEEE International Conference on Data Mining, ICDM 2018
CountrySingapore
CitySingapore
Period11/17/1811/20/18

Fingerprint

Labels
Labeling
Learning systems
Problem-Based Learning
Costs

Keywords

  • Active learning
  • Heterogeneous information networks
  • Multi armed bandit

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Xin, D., El-Kishky, A., Liao, D., Norick, B., & Han, J. (2018). Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach. In 2018 IEEE International Conference on Data Mining, ICDM 2018 (pp. 1350-1355). [8594993] (Proceedings - IEEE International Conference on Data Mining, ICDM; Vol. 2018-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICDM.2018.00184

Active Learning on Heterogeneous Information Networks : A Multi-armed Bandit Approach. / Xin, Doris; El-Kishky, Ahmed; Liao, De; Norick, Brandon; Han, Jiawei.

2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1350-1355 8594993 (Proceedings - IEEE International Conference on Data Mining, ICDM; Vol. 2018-November).

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

Xin, D, El-Kishky, A, Liao, D, Norick, B & Han, J 2018, Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach. in 2018 IEEE International Conference on Data Mining, ICDM 2018., 8594993, Proceedings - IEEE International Conference on Data Mining, ICDM, vol. 2018-November, Institute of Electrical and Electronics Engineers Inc., pp. 1350-1355, 18th IEEE International Conference on Data Mining, ICDM 2018, Singapore, Singapore, 11/17/18. https://doi.org/10.1109/ICDM.2018.00184
Xin D, El-Kishky A, Liao D, Norick B, Han J. Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach. In 2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1350-1355. 8594993. (Proceedings - IEEE International Conference on Data Mining, ICDM). https://doi.org/10.1109/ICDM.2018.00184
Xin, Doris ; El-Kishky, Ahmed ; Liao, De ; Norick, Brandon ; Han, Jiawei. / Active Learning on Heterogeneous Information Networks : A Multi-armed Bandit Approach. 2018 IEEE International Conference on Data Mining, ICDM 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1350-1355 (Proceedings - IEEE International Conference on Data Mining, ICDM).
@inproceedings{b173118fb87f45ec8d81bd1089a94bb8,
title = "Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach",
abstract = "Active learning exploits inherent structures in the unlabeled data to minimize the number of labels required to train an accurate model. It enables effective machine learning in applications with high labeling cost, such as document classification and drug response prediction. We investigate active learning on heterogeneous information networks, with the objective of obtaining accurate node classifications while minimizing the number of labeled nodes. Our proposed algorithm harnesses a multi-armed bandit (MAB) algorithm to determine network structures that identify the most important nodes to the classification task, accounting for node types and without assuming label assortativity. Evaluations on real-world network classification tasks demonstrate that our algorithm outperforms existing methods independent of the underlying classification model.",
keywords = "Active learning, Heterogeneous information networks, Multi armed bandit",
author = "Doris Xin and Ahmed El-Kishky and De Liao and Brandon Norick and Jiawei Han",
year = "2018",
month = "12",
day = "27",
doi = "10.1109/ICDM.2018.00184",
language = "English (US)",
series = "Proceedings - IEEE International Conference on Data Mining, ICDM",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1350--1355",
booktitle = "2018 IEEE International Conference on Data Mining, ICDM 2018",
address = "United States",

}

TY - GEN

T1 - Active Learning on Heterogeneous Information Networks

T2 - A Multi-armed Bandit Approach

AU - Xin, Doris

AU - El-Kishky, Ahmed

AU - Liao, De

AU - Norick, Brandon

AU - Han, Jiawei

PY - 2018/12/27

Y1 - 2018/12/27

N2 - Active learning exploits inherent structures in the unlabeled data to minimize the number of labels required to train an accurate model. It enables effective machine learning in applications with high labeling cost, such as document classification and drug response prediction. We investigate active learning on heterogeneous information networks, with the objective of obtaining accurate node classifications while minimizing the number of labeled nodes. Our proposed algorithm harnesses a multi-armed bandit (MAB) algorithm to determine network structures that identify the most important nodes to the classification task, accounting for node types and without assuming label assortativity. Evaluations on real-world network classification tasks demonstrate that our algorithm outperforms existing methods independent of the underlying classification model.

AB - Active learning exploits inherent structures in the unlabeled data to minimize the number of labels required to train an accurate model. It enables effective machine learning in applications with high labeling cost, such as document classification and drug response prediction. We investigate active learning on heterogeneous information networks, with the objective of obtaining accurate node classifications while minimizing the number of labeled nodes. Our proposed algorithm harnesses a multi-armed bandit (MAB) algorithm to determine network structures that identify the most important nodes to the classification task, accounting for node types and without assuming label assortativity. Evaluations on real-world network classification tasks demonstrate that our algorithm outperforms existing methods independent of the underlying classification model.

KW - Active learning

KW - Heterogeneous information networks

KW - Multi armed bandit

UR - http://www.scopus.com/inward/record.url?scp=85061383211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061383211&partnerID=8YFLogxK

U2 - 10.1109/ICDM.2018.00184

DO - 10.1109/ICDM.2018.00184

M3 - Conference contribution

AN - SCOPUS:85061383211

T3 - Proceedings - IEEE International Conference on Data Mining, ICDM

SP - 1350

EP - 1355

BT - 2018 IEEE International Conference on Data Mining, ICDM 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -