Parameter-setting-free harmony search algorithm

ZW Geem, KB Sim - Applied Mathematics and Computation, 2010 - Elsevier
Applied Mathematics and Computation, 2010Elsevier
Various phenomenon-mimicking algorithms, such as genetic algorithm, simulated
annealing, tabu search, ant colony optimization, and particle swarm optimization, have their
own algorithm parameters. These parameters need to be skillfully assigned in order to
obtain good results. It is burdensome, especially to novice users, to assign these
parameters. The same is true for the harmony search algorithm which was inspired by music
performance. Thus, this study proposes a novel technique to eliminate tedious and …
Various phenomenon-mimicking algorithms, such as genetic algorithm, simulated annealing, tabu search, ant colony optimization, and particle swarm optimization, have their own algorithm parameters. These parameters need to be skillfully assigned in order to obtain good results. It is burdensome, especially to novice users, to assign these parameters. The same is true for the harmony search algorithm which was inspired by music performance. Thus, this study proposes a novel technique to eliminate tedious and experience-requiring parameter assigning efforts. The new parameter-setting-free (PSF) technique which this study suggests contains one additional matrix which contains an operation type (random selection, memory consideration, or pitch adjustment) for every variable in harmony memory. Three examples illustrate that the PSF technique can find good solutions robustly.
Elsevier
以上显示的是最相近的搜索结果。 查看全部搜索结果