Blind multi-class ensemble learning with dependent classifiers

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

1 Citation (Scopus)

Abstract

In recent years, advances in pattern recognition and data analytics have spurred the development of a plethora of machine learning algorithms and tools. However, as each algorithm exhibits different behavior for different types of data, one is motivated to judiciously fuse multiple algorithms in order to find the “best” performing one, for a given dataset. Ensemble learning aims to create such a high-performance meta-learner, by combining the outputs from multiple algorithms. The present work introduces a simple blind scheme for learning from ensembles of classifiers. Blind refers to the combiner who has no knowledge of the ground-truth labels that each classifier has been trained on. While most current works presume that all classifiers are independent, this work introduces a scheme that can handle dependencies between classifiers. Preliminary tests on synthetic data showcase the potential of the proposed approach.

Original languageEnglish (US)
Title of host publication2018 26th European Signal Processing Conference, EUSIPCO 2018
PublisherEuropean Signal Processing Conference, EUSIPCO
Pages2025-2029
Number of pages5
ISBN (Electronic)9789082797015
DOIs
StatePublished - Nov 29 2018
Event26th European Signal Processing Conference, EUSIPCO 2018 - Rome, Italy
Duration: Sep 3 2018Sep 7 2018

Publication series

NameEuropean Signal Processing Conference
Volume2018-September
ISSN (Print)2219-5491

Other

Other26th European Signal Processing Conference, EUSIPCO 2018
CountryItaly
CityRome
Period9/3/189/7/18

Fingerprint

Classifiers
Electric fuses
Learning algorithms
Pattern recognition
Learning systems
Labels

Keywords

  • Dependent classifiers
  • Ensemble learning
  • Multi-class classification
  • Unsupervised

Cite this

Traganitis, P. A., & Giannakis, G. B. (2018). Blind multi-class ensemble learning with dependent classifiers. In 2018 26th European Signal Processing Conference, EUSIPCO 2018 (pp. 2025-2029). [8553113] (European Signal Processing Conference; Vol. 2018-September). European Signal Processing Conference, EUSIPCO. https://doi.org/10.23919/EUSIPCO.2018.8553113

Blind multi-class ensemble learning with dependent classifiers. / Traganitis, Panagiotis A.; Giannakis, Georgios B.

2018 26th European Signal Processing Conference, EUSIPCO 2018. European Signal Processing Conference, EUSIPCO, 2018. p. 2025-2029 8553113 (European Signal Processing Conference; Vol. 2018-September).

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

Traganitis, PA & Giannakis, GB 2018, Blind multi-class ensemble learning with dependent classifiers. in 2018 26th European Signal Processing Conference, EUSIPCO 2018., 8553113, European Signal Processing Conference, vol. 2018-September, European Signal Processing Conference, EUSIPCO, pp. 2025-2029, 26th European Signal Processing Conference, EUSIPCO 2018, Rome, Italy, 9/3/18. https://doi.org/10.23919/EUSIPCO.2018.8553113
Traganitis PA, Giannakis GB. Blind multi-class ensemble learning with dependent classifiers. In 2018 26th European Signal Processing Conference, EUSIPCO 2018. European Signal Processing Conference, EUSIPCO. 2018. p. 2025-2029. 8553113. (European Signal Processing Conference). https://doi.org/10.23919/EUSIPCO.2018.8553113
Traganitis, Panagiotis A. ; Giannakis, Georgios B. / Blind multi-class ensemble learning with dependent classifiers. 2018 26th European Signal Processing Conference, EUSIPCO 2018. European Signal Processing Conference, EUSIPCO, 2018. pp. 2025-2029 (European Signal Processing Conference).
@inproceedings{111f80137c614189865061841a760752,
title = "Blind multi-class ensemble learning with dependent classifiers",
abstract = "In recent years, advances in pattern recognition and data analytics have spurred the development of a plethora of machine learning algorithms and tools. However, as each algorithm exhibits different behavior for different types of data, one is motivated to judiciously fuse multiple algorithms in order to find the “best” performing one, for a given dataset. Ensemble learning aims to create such a high-performance meta-learner, by combining the outputs from multiple algorithms. The present work introduces a simple blind scheme for learning from ensembles of classifiers. Blind refers to the combiner who has no knowledge of the ground-truth labels that each classifier has been trained on. While most current works presume that all classifiers are independent, this work introduces a scheme that can handle dependencies between classifiers. Preliminary tests on synthetic data showcase the potential of the proposed approach.",
keywords = "Dependent classifiers, Ensemble learning, Multi-class classification, Unsupervised",
author = "Traganitis, {Panagiotis A.} and Giannakis, {Georgios B}",
year = "2018",
month = "11",
day = "29",
doi = "10.23919/EUSIPCO.2018.8553113",
language = "English (US)",
series = "European Signal Processing Conference",
publisher = "European Signal Processing Conference, EUSIPCO",
pages = "2025--2029",
booktitle = "2018 26th European Signal Processing Conference, EUSIPCO 2018",

}

TY - GEN

T1 - Blind multi-class ensemble learning with dependent classifiers

AU - Traganitis, Panagiotis A.

AU - Giannakis, Georgios B

PY - 2018/11/29

Y1 - 2018/11/29

N2 - In recent years, advances in pattern recognition and data analytics have spurred the development of a plethora of machine learning algorithms and tools. However, as each algorithm exhibits different behavior for different types of data, one is motivated to judiciously fuse multiple algorithms in order to find the “best” performing one, for a given dataset. Ensemble learning aims to create such a high-performance meta-learner, by combining the outputs from multiple algorithms. The present work introduces a simple blind scheme for learning from ensembles of classifiers. Blind refers to the combiner who has no knowledge of the ground-truth labels that each classifier has been trained on. While most current works presume that all classifiers are independent, this work introduces a scheme that can handle dependencies between classifiers. Preliminary tests on synthetic data showcase the potential of the proposed approach.

AB - In recent years, advances in pattern recognition and data analytics have spurred the development of a plethora of machine learning algorithms and tools. However, as each algorithm exhibits different behavior for different types of data, one is motivated to judiciously fuse multiple algorithms in order to find the “best” performing one, for a given dataset. Ensemble learning aims to create such a high-performance meta-learner, by combining the outputs from multiple algorithms. The present work introduces a simple blind scheme for learning from ensembles of classifiers. Blind refers to the combiner who has no knowledge of the ground-truth labels that each classifier has been trained on. While most current works presume that all classifiers are independent, this work introduces a scheme that can handle dependencies between classifiers. Preliminary tests on synthetic data showcase the potential of the proposed approach.

KW - Dependent classifiers

KW - Ensemble learning

KW - Multi-class classification

KW - Unsupervised

UR - http://www.scopus.com/inward/record.url?scp=85059800211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059800211&partnerID=8YFLogxK

U2 - 10.23919/EUSIPCO.2018.8553113

DO - 10.23919/EUSIPCO.2018.8553113

M3 - Conference contribution

AN - SCOPUS:85059800211

T3 - European Signal Processing Conference

SP - 2025

EP - 2029

BT - 2018 26th European Signal Processing Conference, EUSIPCO 2018

PB - European Signal Processing Conference, EUSIPCO

ER -