当前位置: X-MOL 学术Memetic Comp. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
The discovery of population interaction with a power law distribution in brain storm optimization
Memetic Computing ( IF 3.3 ) Pub Date : 2017-12-28 , DOI: 10.1007/s12293-017-0248-z
Yirui Wang , Shangce Gao , Yang Yu , Zhe Xu

Brain storm optimization (BSO) is a novel evolutionary algorithm which originates from the human brainstorming process. The successful applications of BSO on various problems demonstrate its validity and efficiency. To theoretically analyze the performance of algorithm from the viewpoint of population evolution, the population interaction network (PIN) is used to construct the relationship among individuals in BSO. Four experiments in different dimensions, parameters, combinatorial parameter settings and related algorithms are implemented, respectively. The experimental results indicate the frequency of average degree of BSO meets a power law distribution in the functions with low dimension, which shows the best performance of algorithm among three kinds of dimensions. The parameters of BSO are investigated to find the influence of the population interaction with the power law distribution on the performance of algorithm, and respective parameter can change the relationship among individuals. In addition, the mutual effect among parameters is analyzed to find the best combinatorial result to significantly enhance the performance of BSO. The contrast among BSO, DE and PSO demonstrates a power law distribution is more effective for boosting the population interaction to enhance the performance of algorithm.

中文翻译:

头脑风暴优化中具有幂律分布的种群交互作用的发现

头脑风暴优化(BSO)是一种新颖的进化算法,起源于人类头脑风暴过程。BSO在各种问题上的成功应用证明了其有效性和有效性。从人口进化的角度,从人口交互网络(PIN)理论上分析算法的性能)用于构建BSO中个人之间的关系。分别进行了四个不同维度,参数,组合参数设置和相关算法的实验。实验结果表明,BSO的平均度频率在低维函数中满足幂律分布,表明算法在三种维数中的最佳性能。研究了BSO的参数,以发现人口互动与幂律分布对算法性能的影响,并且各个参数可以改变个体之间的关系。此外,分析了参数之间的相互影响,以找到最佳组合结果,从而显着提高了BSO的性能。BSO之间的对比,
更新日期:2017-12-28
down
wechat
bug