李若愚,
覃华,,
陈琴
广西大学计算机与电子信息学院 南宁 530004
基金项目:国家自然科学基金(61762009)
详细信息
作者简介:苏一丹：男，1962年生，教授，研究方向为自然计算、数据挖掘
李若愚：男，1989年生，硕士生，研究方向为智能优化及在数据挖掘中的应用
覃华：男，1972年生，教授，研究方向为量子计算理论与近似动态规划最优化方法、数据挖掘
陈琴：女，1974年生，副教授，研究方向为数据挖掘
通讯作者:覃华 1694520545@qq.com
中图分类号:TP181计量
文章访问数:1044
HTML全文浏览量:408
PDF下载量:27
被引次数:0
出版历程
收稿日期:2017-11-24
修回日期:2018-04-23
网络出版日期:2018-06-07
刊出日期:2018-08-01
Kernel Extreme Learning Machine Based on K Interpolation Simplex Method
Yidan SU,Ruoyu LI,
Hua QIN,,
Qin CHEN
College of Computer and Electronic Information, Guangxi University, Nanning 530004, China
Funds:The National Natural Science Foundation of China (61762009)
摘要
摘要:针对核极限学习机高斯核函数参数选优难，影响学习机训练收敛速度和分类精度的问题，该文提出一种K插值单纯形法的核极限学习机算法。把核极限学习机的训练看作一个无约束优化问题，在训练迭代过程中，用Nelder-Mead单纯形法搜索高斯核函数的最优核参数，提高所提算法的分类精度。引入K插值为Nelder-Mead单纯形法提供合适的初值，减少单纯形法的迭代次数，提高了新算法的训练收敛效率。通过在UCI数据集上的仿真实验并与其它算法比较，新算法具有更快的收敛速度和更高的分类精度。
关键词:核极限学习机/
核参数/
Nelder-Mead单纯形法/
K插值法
Abstract:The kernel Extreme Learning Machine (ELM) has a problem that the kernel parameter of the Gauss kernel function is hard to be optimized. As a result, training speed and classification accuracy of kernel ELM are negatively affected. To deal with that problem, a novel kernel ELM based on K interpolation simplex method is proposed. The training process of kernel ELM is considered as an unconstrained optimal problem. Then, the Nelder-Mead Simplex Method (NMSM) is used as an optimal method to search the optimized kernel parameter, which improves the classification accuracy of kernel ELM. Furthermore, the K interpolation method is used to provide appropriate initial values for the Nelder-Mead simplex to reduce the number of iterations, and as a result, the training speed of ELM is improved. Comparative results on UCI dataset demonstrate that the novel ELM algorithm has better training speed and higher classification accuracy.
Key words:Kernel Extreme Learning Machine (KELM)/
Kernel parameter/
Nelder-Mead Simplex Method (NMSM)/
K Interpolation method
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=10fed2f4-aeee-465c-860e-610ac3fcebb1
