Training extreme learning machine via regularized correntropy criterion

Training extreme learning machine via regularized correntropy criterion
复制标题

通过正则化熵准则训练极限学习机

DOI:
10.1007/s00521-012-1184-y
复制
发表时间:
2013-12
期刊:
Neural Computing & Applications
影响因子:
--
通讯作者:
Xinmei Wang
Xinmei Wang
中科院分区:
其他
文献类型:
--
作者:
Hongjie Xing;Xinmei Wang

文献摘要

参考文献

被引文献

相似文献

本文提出了一种极限学习机(ELM)的正则化相关熵准则(RCC)来处理含有噪声或异常值的训练集。在RCC中,利用高斯核函数来代替均方误差(MSE)准则的欧几里得范数。用RCC代替MSE可以增强ELM的抗噪声能力。此外,通过半二次(HQ)优化技术以迭代方式可以迅速获得连接隐藏层和输出层的最佳权重以及最佳偏置项。在四个合成数据集和十四个基准数据集上的实验结果表明,该方法优于传统的 ELM 和均由 MSE 准则训练的正则化 ELM。
In this paper, a regularized correntropy criterion (RCC) for extreme learning machine (ELM) is proposed to deal with the training set with noises or outliers. In RCC, the Gaussian kernel function is utilized to substitute Euclidean norm of the mean square error (MSE) criterion. Replacing MSE by RCC can enhance the anti-noise ability of ELM. Moreover, the optimal weights connecting the hidden and output layers together with the optimal bias terms can be promptly obtained by the half-quadratic (HQ) optimization technique with an iterative manner. Experimental results on the four synthetic data sets and the fourteen benchmark data sets demonstrate that the proposed method is superior to the traditional ELM and the regularized ELM both trained by the MSE criterion.
DOI: 10.1016/j.neucom.2010.11.035
发表时间: 2011-09
期刊: Neurocomputing
影响因子: 6
作者:
Jing Deng;Kang Li;G. Irwin
通讯作者: Jing Deng;Kang Li;G. Irwin
DOI: --
发表时间: 1954
期刊: --
影响因子: --
作者:
J. Huseynov
通讯作者: J. Huseynov
DOI: 10.1016/j.neucom.2005.12.126
发表时间: 2006-12-01
期刊: NEUROCOMPUTING
影响因子: 6
作者:
Huang, Guang-Bin;Zhu, Qin-Yu;Siew, Chee-Kheong
通讯作者: Siew, Chee-Kheong
DOI: 10.1007/s13042-011-0049-5
发表时间: 2011-09
影响因子: 5.6
作者:
Binu P. Chacko;V. Krishnan;G. Raju;P. B. Anto
通讯作者: Binu P. Chacko;V. Krishnan;G. Raju;P. B. Anto
DOI: 10.1016/s0165-0114(97)00300-x
发表时间: 1999-09
期刊: Fuzzy Sets Syst.
影响因子: --
作者:
Xizhao Wang;J. Hong
通讯作者: Xizhao Wang;J. Hong