• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>ICCCI 2010;International conference on computer and computational intelligence >BPNN Training based on EPUS-PSO Algorithm
【24h】

BPNN Training based on EPUS-PSO Algorithm

机译：基于EPUS-PSO算法的BPNN训练

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this paper, efficient population utilization strategy for particle swarm optimizer (EPUS-PSO) is introduced to optimize the weight value and threshold value of BP neural network. EPUSPSO is combined with a population manager to omit redundant particles and create new ones or maintain particle numbers according to the solution searching status to make the process more efficient. The algorithm also has two sharing strategies which can stop the particles from falling into the local minimum and make the global optimal solution easier found by particles. Experiments were conducted when the number of hidden nodes is different. EPUS-PSObased BP training mean square error is compared with BP training. The results demonstrate that EPUS-PSObased BP training is superior to BP training.
机译：本文介绍了粒子群优化器（EPUS-PSO）的有效种群利用策略，以优化BP神经网络的权重值和阈值。 EPUSPSO与种群管理器结合使用，可以根据解决方案的搜索状态省略多余的粒子并创建新的粒子或维护粒子数，以使过程更高效。该算法还有两种共享策略，可以阻止粒子陷入局部最小值，并使粒子更容易找到全局最优解。当隐藏节点的数量不同时，进行了实验。将基于EPUS-PSO的BP训练均方误差与BP训练进行比较。结果表明，基于EPUS-PSO的BP训练优于BP训练。

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