Creator |
|
|
Language |
|
Publisher |
|
|
Date |
|
Source Title |
|
First Page |
|
Last Page |
|
Publication Type |
|
Access Rights |
|
Related DOI |
|
|
|
Related URI |
|
|
|
Relation |
|
|
|
Abstract |
We propose a niching method for solving multi-modal optimization problems from their frequency information and evaluate its performance with six multi-modal benchmark functions. We use Fourier transfo...rm as the main analysis tool to obtain the frequency information and estimate the potential modal locations of the fitness landscape using the obtained frequency and phase information. We choose the estimated potential modal locations as elite points, directly put them into the population, and find more peaks within local search areas restricted by given search radiuses. Experimental evaluation results show that our proposed Fourier niching method has the better performance, especially from the computational cost point of view.show more
|