Journal of Computer Applications ›› 2016, Vol. 36 ›› Issue (4): 923-926.DOI: 10.11772/j.issn.1001-9081.2016.04.0923

Previous Articles     Next Articles

Error back propagation algorithm based on iterative MapReduce framework

ZHAO Hu, YANG Yu   

  1. Department of Information Engineering, Engineering University of CAPF, Xi'an Shaanxi 710086, China
  • Received:2015-09-21 Revised:2015-11-25 Online:2016-04-10 Published:2016-04-08
  • Supported by:
    This work is partially supported by the National Natural Science Foundation of China (50575148), the Basic Research Foundation of Engineering College of CAPF(WJY201307), the Basic Research Foundation of Communication Engineering Department of Engineering University of CAPF (XJY201405).

基于迭代式MapReduce的误差反向传播算法

赵虎, 杨宇   

  1. 武警工程大学 信息工程系, 西安 710086
  • 通讯作者: 杨宇
  • 作者简介:赵虎(1991-),男,安徽合肥人,硕士研究生,主要研究向:故障诊断、神经网络、云计算; 杨宇(1981-),男,内蒙古赤峰人,讲师,博士,主要研究方向:故障诊断、神经网络、数据挖掘、隐私保护、云计算。
  • 基金资助:
    国家自然科学基金资助项目(50575148);武警工程大学基础研究基金资助项目(WJY201307);武警工程大学信息工程系基础研究基金资助项目(XJY201405)。

Abstract: Error Back Propagation (BP) algorithm is iterative. How to implement it using iterative MapReduce framework was studied. To avoid the shortage of the traditional MapReduce framework that task must submit repeatedly in iterative program, a transmitting module was added into the iterative MapReduce framework. The training samples obtained from the simulation of the control system in the K/TGR146 radio switch were trained on Hadoop using the traditional framework and the iterative framework respectively. The training of BP algorithm based on the iterative framework is more than 10 times faster than BP algorithm based on the traditional framework, and its accuracy raises by 10%-13%. The iterative framework can effectively reduce the training time and avoid submitting task repeatedly in iterative calculation.

Key words: MapReduce, error Back Propagation (BP) algorithm, iteration, cloud computing

摘要: 针对误差反向传播(BP)算法计算迭代的特点,给出了迭代式MapReduce框架实现BP算法的方法。迭代式MapReduce框架在传统MapReduce框架上添加了传送模块,避免了传统框架运用在迭代程序时需要多次任务提交的缺陷。通过对K/TGR146对空台射电开关控制系统进行仿真得到BP算法训练样本,并在Hadoop云计算环境下,分别在基于传统框架和迭代式框架的BP算法中进行训练。实验结果表明,基于迭代式MapReduce框架的BP算法训练速度达到了基于传统MapReduce框架的BP算法训练速度的10倍以上,正确率提升了10%~13%,能有效解决算法训练时间过长和迭代计算中多次任务提交的问题。

关键词: MapReduce, 误差反向传播算法, 迭代, 云计算

CLC Number: