Implementation and comparison of SVM-based Multi-Task Learning methods

Han Tai Shiao; Vladimir S Cherkassky

doi:10.1109/IJCNN.2012.6252442

Implementation and comparison of SVM-based Multi-Task Learning methods

Han Tai Shiao, Vladimir S Cherkassky

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

6 Scopus citations

Abstract

Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised-learning applications, data can be naturally separated into several groups, or tasks, and incorporating this information into learning may improve generalization. There are many Multi-Task Learning (MTL) techniques for classification recently proposed in machine learning. This paper focuses on analysis and comparison of the two recent SVM-based MTL techniques: regularized MTL (rMTL) and SVM+ based MTL (SVM+MTL). In particular, our analysis shows how these two methods can be implemented using standard SVM software. Further, we present extensive empirical comparisons between these two methods, which relates advantages/limitations of each method to statistical characteristics of the training data.

Original language	English (US)
Title of host publication	2012 International Joint Conference on Neural Networks, IJCNN 2012
DOIs	https://doi.org/10.1109/IJCNN.2012.6252442
State	Published - 2012
Event	2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012 - Brisbane, QLD, Australia Duration: Jun 10 2012 → Jun 15 2012

Publication series

Name	Proceedings of the International Joint Conference on Neural Networks

Other

Other	2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012
Country/Territory	Australia
City	Brisbane, QLD
Period	6/10/12 → 6/15/12

Keywords

SVM-Plus (SVM+)
classification
land mine data
model selection
multi-task learning (MTL)
support vector machine

Access

10.1109/IJCNN.2012.6252442

OpenUrl availability

Full text

Cite this

Shiao, HT & Cherkassky, VS 2012, Implementation and comparison of SVM-based Multi-Task Learning methods. in 2012 International Joint Conference on Neural Networks, IJCNN 2012., 6252442, Proceedings of the International Joint Conference on Neural Networks, 2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012, Brisbane, QLD, Australia, 6/10/12. https://doi.org/10.1109/IJCNN.2012.6252442

@inproceedings{342f6c0751374dd2ad798688f6ce9a8c,

title = "Implementation and comparison of SVM-based Multi-Task Learning methods",

abstract = "Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised-learning applications, data can be naturally separated into several groups, or tasks, and incorporating this information into learning may improve generalization. There are many Multi-Task Learning (MTL) techniques for classification recently proposed in machine learning. This paper focuses on analysis and comparison of the two recent SVM-based MTL techniques: regularized MTL (rMTL) and SVM+ based MTL (SVM+MTL). In particular, our analysis shows how these two methods can be implemented using standard SVM software. Further, we present extensive empirical comparisons between these two methods, which relates advantages/limitations of each method to statistical characteristics of the training data.",

keywords = "SVM-Plus (SVM+), classification, land mine data, model selection, multi-task learning (MTL), support vector machine",

author = "Shiao, {Han Tai} and Cherkassky, {Vladimir S}",

year = "2012",

doi = "10.1109/IJCNN.2012.6252442",

language = "English (US)",

isbn = "9781467314909",

series = "Proceedings of the International Joint Conference on Neural Networks",

booktitle = "2012 International Joint Conference on Neural Networks, IJCNN 2012",

note = "2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012 ; Conference date: 10-06-2012 Through 15-06-2012",

}

TY - GEN

T1 - Implementation and comparison of SVM-based Multi-Task Learning methods

AU - Shiao, Han Tai

AU - Cherkassky, Vladimir S

PY - 2012

Y1 - 2012

N2 - Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised-learning applications, data can be naturally separated into several groups, or tasks, and incorporating this information into learning may improve generalization. There are many Multi-Task Learning (MTL) techniques for classification recently proposed in machine learning. This paper focuses on analysis and comparison of the two recent SVM-based MTL techniques: regularized MTL (rMTL) and SVM+ based MTL (SVM+MTL). In particular, our analysis shows how these two methods can be implemented using standard SVM software. Further, we present extensive empirical comparisons between these two methods, which relates advantages/limitations of each method to statistical characteristics of the training data.

AB - Exploiting additional information to improve traditional inductive learning is an active research area in machine learning. In many supervised-learning applications, data can be naturally separated into several groups, or tasks, and incorporating this information into learning may improve generalization. There are many Multi-Task Learning (MTL) techniques for classification recently proposed in machine learning. This paper focuses on analysis and comparison of the two recent SVM-based MTL techniques: regularized MTL (rMTL) and SVM+ based MTL (SVM+MTL). In particular, our analysis shows how these two methods can be implemented using standard SVM software. Further, we present extensive empirical comparisons between these two methods, which relates advantages/limitations of each method to statistical characteristics of the training data.

KW - SVM-Plus (SVM+)

KW - classification

KW - land mine data

KW - model selection

KW - multi-task learning (MTL)

KW - support vector machine

UR - http://www.scopus.com/inward/record.url?scp=84865084374&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865084374&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2012.6252442

DO - 10.1109/IJCNN.2012.6252442

M3 - Conference contribution

AN - SCOPUS:84865084374

SN - 9781467314909

T3 - Proceedings of the International Joint Conference on Neural Networks

BT - 2012 International Joint Conference on Neural Networks, IJCNN 2012

T2 - 2012 Annual International Joint Conference on Neural Networks, IJCNN 2012, Part of the 2012 IEEE World Congress on Computational Intelligence, WCCI 2012

Y2 - 10 June 2012 through 15 June 2012

ER -

Implementation and comparison of SVM-based Multi-Task Learning methods

Abstract

Publication series

Other

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this