Effectively training neural machine translation models with monolingual data

doi:10.1016/j.neucom.2018.12.032

	Effectively training neural machine translation models with monolingual data
	Yang, Zhen1,2 ; Chen, Wei2 ; Wang, Feng2 ; Xu, Bo2
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
	2019-03-14
卷号	333 页码:240-247
通讯作者	Yang, Zhen(yangzhen2014@ia.ac.cn)
摘要	Improving neural machine translation models (NMT) with monolingual data has aroused more and more interests in this area and back-translation for monolingual data augmentation Sennrich et al. (2016) has been taken as a promising development recently. While the naive back-translation approach improves the translation performance substantially, we notice that its usage for monolingual data is not so effective because traditional NMT models make no distinction between the true parallel corpus and the back translated synthetic parallel corpus. This paper proposes a gate-enhanced NMT model which makes use of monolingual data more effectively. The central idea is to separate the data flow of monolingual data and parallel data into different channels by the elegant designed gate, which enables the model to perform different transformations according to the type of the input sequence, i.e., monolingual data and parallel data. Experiments on Chinese-English and English-German translation tasks show that our approach achieves substantial improvements over strong baselines and the gate-enhanced NMT model can utilize the source-side and target-side monolingual data at the same time. (C) 2018 Elsevier B.V. All rights reserved.
关键词	Neural machine translation Monolingual data Gate-enhanced Source-side and target-side Effectively
DOI	10.1016/j.neucom.2018.12.032
关键词[WOS]	NETWORK
收录类别	SCI
语种	英语
资助项目	National Program on Key Basic Research Project of China (973 Program)[2013CB329302] ; National Program on Key Basic Research Project of China (973 Program)[2013CB329302]
项目资助者	National Program on Key Basic Research Project of China (973 Program)
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000456834100022
出版者	ELSEVIER SCIENCE BV
七大方向——子方向分类	自然语言处理
引用统计	被引频次：10[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/25314
专题	复杂系统认知与决策实验室_听觉模型与认知计算
通讯作者	Yang, Zhen
作者单位	1.Univ Chinese Acad Sci, Beijing, Peoples R China 2.Chinese Acad Sci, Inst Automat, 95 ZhongGuanCun East Rd, Beijing 100190, Peoples R China
第一作者单位	中国科学院自动化研究所
通讯作者单位	中国科学院自动化研究所
推荐引用方式 GB/T 7714	Yang, Zhen,Chen, Wei,Wang, Feng,et al. Effectively training neural machine translation models with monolingual data[J]. NEUROCOMPUTING,2019,333:240-247.
APA	Yang, Zhen,Chen, Wei,Wang, Feng,&Xu, Bo.(2019).Effectively training neural machine translation models with monolingual data.NEUROCOMPUTING,333,240-247.
MLA	Yang, Zhen,et al."Effectively training neural machine translation models with monolingual data".NEUROCOMPUTING 333(2019):240-247.