Kagan Tumer's Publications

Display Publications by [Year] [Type] [Topic]


Order Statistics Combiners for Neural Classifiers. K. Tumer and J. Ghosh. In Proceedings of the World Congress on Neural Networks, pp. I:31–34, INNS Press, Washington D.C., July 1995.

Abstract

Several researchers have shown that linearly combining outputs of multiple neural classifiers results in better performance for many applications. In this paper we introduce a family of order statistics combiners as an alternative to linear combiners. We show analytically that the selection of the median, the maximum and in general, the ith order statistic improves classification performance. Specifically, we show that order statistics combiners reduce the variance of the actual decision boundaries around the optimum boundary, and that this is directly related to classification error.

Download

(unavailable)

BibTeX Entry

@inproceedings{tumer-ghosh_wcnn95,
        author={K. Tumer and J. Ghosh},
        title="Order Statistics Combiners for Neural Classifiers",
        booktitle="Proceedings of the World Congress on Neural Networks",
        publisher="{INNS} Press",
        pages ={I:31-34},
        address="Washington D.C.",
	month ={July},
	abstract={Several researchers have shown that linearly combining outputs of multiple neural classifiers results in better performance for many applications. In this paper we introduce a family of order statistics combiners as an alternative to linear combiners. We show analytically that the selection of the median, the maximum and in general, the ith order statistic improves classification performance. Specifically, we show that order statistics combiners reduce the variance of the actual decision boundaries around the optimum boundary, and that this is directly related to classification error.},
	bib2html_pubtype = {Refereed Conference Papers},
	bib2html_rescat = {Classifier Ensembles},
        year={1995}
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Apr 01, 2020 17:39:43