mk:kes03

Summary

On the Benchmarking of Multiobjective Optimization Algorithm. Mario Köppen. In Knowledge-Based Intelligent Information and Engineering Systems (KES 2003), Proceedings, pages 379-385, 2003. (URL)

Abstract

The "No Free Lunch" (NFL) theorems state that in average each algorithm has the same performance, when no a priori knowledge of single-objective cost function f is assumed. This paper extends the NFL theorems to the case of multi-objective optimization. Further it is shown that even in cases of a priori knowledge, when the performance measure is related to the set of extrema points sampled so far, the NFL theorems still hold. However, a procedure for obtaining function-dependent algorithm performance can be constructed, the so-called tournament performance, which is able to gain different performance measures for different multiobjective algorithms.

Bibtex entry

@INPROCEEDINGS { mk:kes03,
    ABSTRACT = { The "No Free Lunch" (NFL) theorems state that in average each algorithm has the same performance, when no a priori knowledge of single-objective cost function f is assumed. This paper extends the NFL theorems to the case of multi-objective optimization. Further it is shown that even in cases of a priori knowledge, when the performance measure is related to the set of extrema points sampled so far, the NFL theorems still hold. However, a procedure for obtaining function-dependent algorithm performance can be constructed, the so-called tournament performance, which is able to gain different performance measures for different multiobjective algorithms. },
    AUTHOR = { Mario Köppen },
    BOOKTITLE = { Knowledge-Based Intelligent Information and Engineering Systems (KES 2003), Proceedings },
    MODIFIED = { 2008-02-28 14:37:56 +0900 },
    EDITOR = { Vasile Palade and Robert J. Howlett and Lakhmi Jain },
    HASABSTRACT = { Yes },
    PAGES = { 379-385 },
    PDF = { kes03.pdf },
    PUBLISHER = { Springer-Verlag Heidelberg },
    SERIES = { LNAI 2773 },
    TITLE = { On the Benchmarking of Multiobjective Optimization Algorithm },
    URL = { http://www.springerlink.com/content/kl0bvhj8qhr1batg },
    YEAR = { 2003 },
    1 = { http://www.springerlink.com/content/kl0bvhj8qhr1batg },
}

On small computer displays, you can hide this right bar by using the 'Hide' button above.

News

Next conferences COMPSAC 2014 (Vasteras, Sweden, July 2014), INCoS-2014 (Salerno, Italy, September 2014).

New edited book "Soft Computing in Industrial Applications", V. Snasel, P. Kroemer, M. Koeppen, G. Schaefer, Springer AISC 223, July 2013.