| 本文已被：浏览 4760次 下载 6936次 |
码上扫一扫！ |
|
|
| 鲁棒性的模糊聚类神经网络 |
|
邓赵红1,2,3, 王士同1,2,3
|
|
1.江南大学,信息工程学院,江苏,无锡,214036;2.计算机软件新技术国家重点实验室(南京大学),江苏,南京,210093;3.中国科学院,计算机科学重点实验室,北京,100080
|
|
| 摘要: |
| 针对模糊聚类神经网络FCNN(fuzzy clustering neural network)对例外点(outliers)敏感的缺陷,通过引入Vapnik's ε-不敏感损失函数,重新构造网络的目标函数,并根据拉格朗日优化理论推导出新的鲁棒模糊聚类神经网络及其算法(robust fuzzy clustering neural networks,简称RFCNN).RFCNN有效地克服了FCNN对例外点敏感之缺点并且能得到合理的聚类中心.仿真实验结果表明,RFCNN较之于FCNN有更好的鲁棒性. |
| 关键词: 模糊聚类 神经网络 ε-不敏感损失函数 例外点 鲁棒性 |
| DOI： |
| 分类号: |
| 基金项目:Supported by the National Natural Science Foundation of China under Grant No.60225015(国家自然科学基金);the Natural Science Foundation of Jiangsu Province of China under Grant No.BK2003017(江苏省自然科学基金);the 2004-New-Century Outstanding Young Teacher Grant of Ministry of Education of China under Grant No.NCET-040496(2004年度国家教育部新世纪优秀人才计划);the 2005 Key Research Project of Ministry of Education of China(2005年度国家教育部科学研究重点项目) |
|
| Robust Fuzzy Clustering Neural Networks |
|
DENG Zhao-Hong,WANG Shi-Tong
|
| Abstract: |
| In this paper a new robust fuzzy clustering neural networks (RFCNN) is presented to resolve the sensitivity of the fuzzy clustering neural network (FCNN) to outliers in real datasets. The new objective function of RFCNN is obtained by introducing Vapnik’s ε-insensitive loss function, and RFCNN’s update rules are derived by using Lagrange optimization theory. Compared with the FCNN algorithm, RFCNN is much more robust to outliers in the datasets. Experimental results demonstrate the effectiveness of RFCNN. |
| Key words: fuzzy clustering neural network ε-insensitive loss function outliers robustness |
