计算机应用研究2016，Vol.33Issue(3)：843-847,5.DOI:10.3969/j.issn.1001-3695.2016.03.046
一种交替变换更新层数的 DBN-DNN 快速训练方法
Alternating update layers for DBN-DNN fast training method
李轩 1李春升1
作者信息
- 1. 北京航空航天大学 电子信息工程学院，北京 100191
- 折叠
摘要
Abstract
As for the problem of too long training time of DBN-DNN in speech recognition,this paper proposed a fast training method to solve it.From the view of reducing the error back propagation calculation,this method achieved acceleration by alter-nating update layers.In the paper,it also designed two kinds of implementation strategy,i.e.shrinking global update frequency (SGUF)and shrinking partial update layer(SPUL).The method could be combined with a variety of fast training algorithm for DNN.The experimental results show that by this method independently or by the combination with Stochastic Data Sweeping (SDS)or ASGD algorithm,training time will be reduced dramatically at no loss of recognition accuracy.关键词
语音识别/DBN-DNN/快速训练算法Key words
speech recognition/DBN-DNN/fast training method分类
信息技术与安全科学引用本文复制引用
李轩,李春升..一种交替变换更新层数的 DBN-DNN 快速训练方法[J].计算机应用研究,2016,33(3):843-847,5.
