Open Access. Powered by Scholars. Published by Universities.®
U.C. Berkeley Division of Biostatistics Working Paper Series
Evolutionary computation; performance comparison; two-fold sampling; multiple hypothesis testing; bootstrap; numeric optimization
Articles 1 - 1 of 1
Full-Text Articles in Statistics and Probability
A General Framework For Statistical Performance Comparison Of Evolutionary Computation Algorithms, David Shilane, Jarno Martikainen, Sandrine Dudoit, Seppo Ovaska
A General Framework For Statistical Performance Comparison Of Evolutionary Computation Algorithms, David Shilane, Jarno Martikainen, Sandrine Dudoit, Seppo Ovaska
U.C. Berkeley Division of Biostatistics Working Paper Series
This paper proposes a statistical methodology for comparing the performance of evolutionary computation algorithms. A two-fold sampling scheme for collecting performance data is introduced, and these data are analyzed using bootstrap-based multiple hypothesis testing procedures. The proposed method is sufficiently flexible to allow the researcher to choose how performance is measured, does not rely upon distributional assumptions, and can be extended to analyze many other randomized numeric optimization routines. As a result, this approach offers a convenient, flexible, and reliable technique for comparing algorithms in a wide variety of applications.