《计算机应用研究》|Application Research of Computers

基于资源分配网络的小数据集并行集成学习方法

Parallel ensemble learning method based on resource allocating networks for small dataset

免费全文下载 (已被下载 次)  
获取PDF全文
作者 张安国,张树勋,朱巍,李秀敏,黄金龙
机构 1.锐捷网络股份有限公司 锐捷研究院,福州 350002;2.中国科学院新疆理化技术研究所,乌鲁木齐 830011;3.中国科学院大学,北京 100049;4.新疆民族语音语言信息处理实验室,乌鲁木齐 830011;5.重庆大学 自动化学院,重庆 400044;6.长江师范学院,重庆 408100
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)04-009-0997-04
DOI 10.19734/j.issn.1001-3695.2017.10.0984
摘要 为了在小规模的训练数据集上获得一个具有稳定的高计算精度的算法模型,提出了一种基于扩展卡尔曼滤波器的资源分配网络并行集成学习方法。该集成系统由多个带有扩展卡尔曼滤波器的资源分配网络(RANEKF)组成,并且每个RANEKF子网的输入由原始数据集中的输入经过随机权值的修正得到。通过和其他神经网络构成的集成学习算法的实验对比,发现提出的方法在小训练集上拥有更高的计算精度和稳定性。
关键词 资源分配网络; 并行集成学习; 增量学习; 扩展卡尔曼滤波器
基金项目 重庆市自然科学基金资助项目(2016jcyjA0015)
重庆市涪陵区科技计划项目(FLKJ2015ABB1099)
本文URL http://www.arocmag.com/article/01-2019-04-009.html
英文标题 Parallel ensemble learning method based on resource allocating networks for small dataset
作者英文名 Zhang Anguo, Zhang Shuxun, Zhu Wei, Li Xiumin, Huang Jinlong
机构英文名 1.Research Institute of Ruijie,Ruijie Networks Co,Ltd,Fuzhou 350002,China;2.Xinjiang Technical Institute of Physical & Chemical,Chinese Academy of Sciences,Urumqi 830011,China;3.University of Chinese Academy of Sciences,Beijing 100049,China;4.Xinjiang Laboratory of Minority Speech & Language Information Processing,Urumqi 830011,China;5.College of Automation,Chongqing University,Chongqing 400044,China;6.Yangzte Normal University,Chongqing 408100,China
英文摘要 To design a training model with stable computational performance and high accuracy which is applied on a small training dataset, this paper proposed a resource allocating networks with extended Kalman filter(RANEKF) based parallel ensemble learning algorithm. The learning system is composed of multiple RANEKF units, and the unit inputs are produced by the original dataset with random initialized weights. The experiment results conducted on a small dataset show that the novel model outperforms the ensemble learning systems constructed by the other artificial neural networks in terms of the computational accuracy and stability.
英文关键词 resource allocating network; parallel ensemble learning; incremental learning; extended Kalman filter
参考文献 查看稿件参考文献
 
收稿日期 2017/10/26
修回日期 2017/12/20
页码 997-1000
中图分类号 TP183
文献标志码 A