首页> 外文会议>2017 Intelligent Systems Conference >Cluster restarted DM: New algorithm for global optimisation
【24h】

Cluster restarted DM: New algorithm for global optimisation

机译:集群重启的DM:用于全局优化的新算法

获取原文
获取原文并翻译 | 示例

摘要

Global optimisation method Differential Migration (DM) with restarting is described in this paper and evaluated together with Restart Covariance Matrix Adaptation Evolution Strategy With Increasing Population Size (IPOP-CMA-ES). Differential Migration is another step in global optimisation from SOMA (Self-Organizing Migration Algorithm) combining two basic individual movement methods of SOMA - all-to-one and all-to-all, via cluster analysis and internal algorithm constant defining continuous change from one type of movement to another. The proposed algorithm implements essential ideas of Differential Evolution regardless of their original interpretation in living nature with subsequent increase of efficiency in finding global extreme which holds mainly for noisy multimodal cost functions present in the benchmarks as well as in real world applications.
机译:本文介绍了具有重新启动的全局优化方法(DM),并将其与随着人口规模增加的重新启动协方差矩阵适应进化策略(IPOP-CMA-ES)一起进行了评估。差异迁移是SOMA(自组织迁移算法)进行全局优化的又一步骤,结合了SOMA的两种基本的个体移动方法,即通过集群分析和内部算法常量定义了从一个对象的连续更改的全对一和全对所有方法。运动到另一个类型。所提出的算法实现了“差分进化”的基本思想,而不管它们对生命本质的最初解释如何,随后发现全局极限的效率随之提高,这主要适用于基准以及现实应用中存在的嘈杂多峰成本函数。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号