计算机科学 ›› 2010, Vol. 37 ›› Issue (3): 208-211252.

• 人工智能 • 上一篇    下一篇

结合变邻域搜索的竞争Hopfield神经网络解决最大分散度问题

周雅兰,王甲海,闭玮,莫斌,李曙光   

  1. (广东商学院信息学院 广州510320);(中山大学信息科学与技术学院 广州510006)
  • 出版日期:2018-12-01 发布日期:2018-12-01
  • 基金资助:
    本文受l国家自然科学基金(60805026,60905038),高等学校博十学科点专项科研基金(20070668052,教育部留学回国人员科研启动基金(教外司留[2007]1108号),广东商学院校级科研项目(08BS52001)资助。

Competitive Hopfield Network Combined with Variable Neighborhood Search for Maximum Diversity Problems

ZHOU Ya-lan,WANG Jia-hai,BI Wei,MO Bin,LI Shu-guang   

  • Online:2018-12-01 Published:2018-12-01

摘要: 提出一种结合变部域搜索的离散竞争Hopfield神经网络,用于求解最大分散度问题。为了克服神经网络易陷入局部最小值的问题,将变邻域搜索的思想引入到离散竟争Hopficld神经网络中,一旦网络陷入局部最小值,变邻 域搜索能帮助神经网络动态改变搜索部域,从而跳出局部最小值去搜寻更优的解。最后,针对最大分散度问题的实验结果表明,提出的算法具有良好的性能。

关键词: 变部域搜索,Hopfield神经网络,最大分散度问题

Abstract: A discrete competitive Hopfield neural network(I}CHNN) combinai with variable neighborhood search(VNS) was proposed for the maximum diversity problem In order to overcome the local minima problem of D(}HNN,the idea of VNS was introduced into DCHNN. Once the network is trapped in local minima, the VNS can generate a new search neighborhood for I}CHNN. I}hus, the proposed algorithm can escape from local minima and further search better results. Finally, simulation results on the maximum diversity problem show that the proposed algorithm has good performancc.

Key words: Variable neighborhood search, Hopfield network, Maximum diversity problem

No related articles found!
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!