Title :
Efficiency competition on N-queen problem: DE vs. CMA-ES
Author :
Rahnamayan, Shahryar ; Dieras, Paul
Author_Institution :
Dept. of Mechatron. Syst. Eng., Simon Fraser Univ., Vancouver, BC
Abstract :
In this paper, two well-known evolutionary algorithms, namely, differential evolution (DE) and covariance matrix adaptation evolution strategy (CMA-ES), are compared on solving N-queen problem (a classical combinatorial optimization problem). Convergence velocity and robustness are our main measures in the current comparative study. Experiments are conducted on twelve chessboard dimensions. Results confirm that CMA-ES performs better than DE with respect to success rate and also success performance. Details about the N-queen problem, comparison strategies, metrics, and results are provided.
Keywords :
combinatorial mathematics; covariance matrices; evolutionary computation; N-queen problem; convergence velocity; covariance matrix adaptation evolution strategy; differential evolution; evolutionary algorithms; Convergence; Covariance matrix; Current measurement; Evolutionary computation; Genetic mutations; Mechatronics; Random number generation; Robustness; Systems engineering and theory; Velocity measurement;
Conference_Titel :
Electrical and Computer Engineering, 2008. CCECE 2008. Canadian Conference on
Conference_Location :
Niagara Falls, ON
Print_ISBN :
978-1-4244-1642-4
Electronic_ISBN :
0840-7789
DOI :
10.1109/CCECE.2008.4564490
