首页> 外文OA文献 >Agreement-Based Joint Training for Bidirectional Attention-Based Neural Machine Translation

【2h】

Agreement-Based Joint Training for Bidirectional Attention-Based Neural Machine Translation

机译：基于协议的基于双向关注的神经机翻译的联合培训

代理获取

本网站仅为用户提供外文OA文献查询和代理获取服务，本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文，但由于OA文献来源多样且变更频繁，仍可能出现获取不到、文献不完整或与标题不符等情况，如果获取不到我们将提供退款服务。请知悉。

页面导航

摘要
著录项
相似文献
相关主题

摘要

The attentional mechanism has proven to be effective in improving end-to-endneural machine translation. However, due to the intricate structural divergencebetween natural languages, unidirectional attention-based models might onlycapture partial aspects of attentional regularities. We propose agreement-basedjoint training for bidirectional attention-based end-to-end neural machinetranslation. Instead of training source-to-target and target-to-sourcetranslation models independently,our approach encourages the two complementarymodels to agree on word alignment matrices on the same training data.Experiments on Chinese-English and English-French translation tasks show thatagreement-based joint training significantly improves both alignment andtranslation quality over independent training.

机译：预付机制已被证明是有效地改善终端电机翻译。然而，由于自然语言的复杂结构分歧，基于单向关注的模型可能只能占据注意力的部分方面。我们提出了基于双向关注的端到端神经机械进程的协议培训。我们的方法而不是培训源目标和目标到型号模型，我们的方法鼓励两个互补的模型在同一培训数据上同意词对齐矩阵。论中英文和英语 - 法语翻译任务的分析显示据说基于象征联合培训在独立培训方面显着提高了对准和翻译质量。

著录项

作者
Yong Cheng;
展开▼
作者单位

展开▼
年度 2019
总页数
原文格式 PDF
正文语种
中图分类

相似文献

外文文献
中文文献
专利

1. Exploiting reverse target-side contexts for neural machine translation via asynchronous bidirectional decoding [J] . Su Jinsong, Zhang Xiangwen, Lin Qian, Artificial intelligence . 2019,第Deca期

机译：通过异步双向解码利用反向目标侧上下文进行神经机器翻译
2. Tigrinya Morphological Segmentation with Bidirectional Long Short-Term Memory Neural Networks and Its Effect on English-Tigrinya Machine Translation [J] . Yemane Keleta Tedla 人工知能: 人工知能学会誌 . 2019,第1期

机译：双向短期内记忆神经网络的Tigrinya形态分割及其对英语 - Tigrinya Machine翻译的影响
3. A Joint Back-Translation and Transfer Learning Method for Low-Resource Neural Machine Translation [J] . Gong-Xu Luo, Ya-Ting Yang, Rui Dong, Mathematical Problems in Engineering: Theory, Methods and Applications . 2020,第1期

机译：低资源神经电机翻译的联合背翻译与转移学习方法
4. Agreement-Based Joint Training for Bidirectional Attention-Based Neural Machine Translation [C] . Yong Cheng, Shiqi Shen, Zhongjun He, International Joint Conference on Artificial Intelligence . 2016

机译：基于协议的基于双向关注的神经机翻译的联合培训
5. Machine learning approaches for dealing with limited bilingual training data in statistical machine translation. [D] . Haffari, Gholamreza. 2009

机译：在统计机器翻译中用于处理有限的双语培训数据的机器学习方法。
6. Biomedical word sense disambiguation with bidirectional long short-term memory and attention-based neural networks [O] . Canlin Zhang, Daniel Biś, Xiuwen Liu, 2019

机译：具有双向长期短期记忆和基于注意力的神经网络的生物医学单词义消歧
7. Towards Bidirectional Hierarchical Representations for Attention-Based Neural Machine Translation [O] . Yang, Baosong, Wong, Derek F., Xiao, Tong, 2017

机译：面向注意力的双向层次表示神经机器翻译
8. Odds of Successful Transfer of Low-level Concepts: A Key Metric for Bidirectional Speech-to-Speech Machine Translation in DARPA's TRANSTAC Program [R] . Sanders, G. A., Bronsart, S., Condon, S., 2008

机译：成功转移低级概念的几率：DaRpa TRaNsTaC计划中双向语音到语音机器翻译的关键指标

Agreement-Based Joint Training for Bidirectional Attention-Based Neural Machine Translation

摘要

著录项

相似文献

相关主题

期刊订阅