MULTI-OBJECTIVE EVOLUTION OF THE PARETO OPTIMAL SET OF NEURAL NETWORK CLASSIFIER ENSEMBLES

This source preferred by Keith Phalp

Authors: Engen, V., Vincent, J., Schierz, A.C. and Phalp, K.T.

Start date: 12 July 2009

Pages: 74-79

Publisher: IEEE

DOI: 10.1109/ICMLC.2009.5212485

Existing research demonstrates that classifier ensem- bles can improve on the performance of the single ‘best’ classifier. However, for some problems, although the ensemble may obtain a lower classification error than any of the base classifiers, it may not provide the de- sired trade-off among the classification rates of differ- ent classes. In many applications, classes are not of equal importance, but the preferred trade-off may be hard to quantify a priori. In this paper, we adopt multi- objective techniques to create Pareto optimal sets of classifiers and ensembles, offering the user the choice of preferred trade-off. We also demonstrate that the common practice of developing a single ensemble from an arbitrary (diverse) selection of base classifiers will be inferior to a large proportion of those classifiers.

This data was imported from Scopus:

Authors: Engen, V., Vincent, J., Schierz, A.C. and Phalp, K.

Journal: Proceedings of the 2009 International Conference on Machine Learning and Cybernetics

Volume: 1

Pages: 74-79

ISBN: 9781424437030

DOI: 10.1109/ICMLC.2009.5212485

Existing research demonstrates that classifier ensembles can improve on the performance of the single 'best' classifier. However, for some problems, although the ensemble may obtain a lower classification error than any of the base classifiers, it may not provide the desired trade-off among the classification rates of different classes. In many applications, classes are not of equal importance, but the preferred trade-off may be hard to quantify a priori. In this paper, we adopt multi-objective techniques to create Pareto optimal sets of classifiers and ensembles, offering the user the choice of preferred trade-off. We also demonstrate that the common practice of developing a single ensemble from an arbitrary (diverse) selection of base classifiers will be inferior to a large proportion of those classifiers. © 2009 IEEE.

This data was imported from Web of Science (Lite):

Authors: Engen, V., Vincent, J., Schierz, A.C., Phalp, K. and IEEE

Journal: PROCEEDINGS OF 2009 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-6

Pages: 74-79

ISBN: 978-1-4244-4705-3

DOI: 10.1109/ICMLC.2009.5212485

The data on this page was last updated at 05:02 on September 21, 2018.