|
|||||
|
|
| 改进的遗传谐振匹配网络学习算法 | |
| 引用本文： | 温长吉,王生生,于合龙,苏恒强.改进的遗传谐振匹配网络学习算法[J].华中科技大学学报(自然科学版),2011,39(11):47-49,67. |
| 作者姓名： | 温长吉 王生生 于合龙 苏恒强 |
| 作者单位： | 1. 吉林农业大学信息技术学院,吉林长春130118／吉林大学计算机科学与技术学院,吉林长春130012 2. 吉林大学计算机科学与技术学院,吉林长春,130012 3. 吉林农业大学信息技术学院,吉林长春,130118 |
| 基金项目： | 国家自然科学基金资助项目(60773099,60873149,60973088); 吉林大学科学前沿与交叉学科创新资助项目(200903178) |
| 摘 要： | 针对简单自适应匹配（SFAM）网络结构冗余的问题提出一种改进遗传谐振匹配网络学习算法.通过引入群体适应度均值和标准差来自适应调整交叉概率和变异概率;针对自适应谐振网络自身在训练过程中可能出现的类冗余问题,在遗传操作中引入裁剪算子,通过定义置信度因子依据规则对训练过程中网络可能出现的冗余类标志节点进行删除,降低网络结构的... |
| 关 键 词： | 简单自适应谐振匹配 自适应遗传算法 裁剪算子 置信度因子 过学习 |
Improved genetic resonance matching network learning algorithm |
|
| Wen Changji, Wang Shengsheng Yu Helong Su Hengqiang.Improved genetic resonance matching network learning algorithm[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,2011,39(11):47-49,67. | |
| Authors: | Wen Changji Wang Shengsheng Yu Helong Su Hengqiang |
| Institution: | Wen Changji1,2 Wang Shengsheng2 Yu Helong1 Su Hengqiang1(1 School of Information Technology,Jilin Agriculture University,Changchun 130118,China,2 College of Computer Science and Technology,Jilin University,Changchun 130012,China) |
| Abstract: | Aimed at the structural redundancy of simple fuzzy adaptive matching(SFAM) network,an improved genetic resonance matching network learning algorithm was proposed.By means of population average fitness and standard deviation,crossover probability and mutation probability were adjusted by these strategies adaptively;then,by considering the class redundancy appeared over the self-training of SFAM,cutting operators were proposed to the genetic operations.To reduce the redundancy of network structure,by defining... |
| Keywords: | simple fuzzy adaptive matching(SFAM) self-adaption genetic algorithm cutting operator credibility factors overfitting |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |