A Unified Neural Network for Quality Estimation of Machine Translation

Maoxi LI  Qingyu XIANG  Zhiming CHEN  Mingwen WANG  

Publication
IEICE TRANSACTIONS on Information and Systems   Vol.E101-D   No.9   pp.2417-2421
Publication Date: 2018/09/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2018EDL8019
Type of Manuscript: LETTER
Category: Natural Language Processing
Keyword: 
natural language processing,  machine translation,  neural quality estimation,  recurrent neural network (RNN),  bidirectional RNN encoder-decoder with attention mechanism,  

Full Text: PDF(610.8KB)
>>Buy this Article


Summary: 
The-state-of-the-art neural quality estimation (QE) of machine translation model consists of two sub-networks that are tuned separately, a bidirectional recurrent neural network (RNN) encoder-decoder trained for neural machine translation, called the predictor, and an RNN trained for sentence-level QE tasks, called the estimator. We propose to combine the two sub-networks into a whole neural network, called the unified neural network. When training, the bidirectional RNN encoder-decoder are initialized and pre-trained with the bilingual parallel corpus, and then, the networks are trained jointly to minimize the mean absolute error over the QE training samples. Compared with the predictor and estimator approach, the use of a unified neural network helps to train the parameters of the neural networks that are more suitable for the QE task. Experimental results on the benchmark data set of the WMT17 sentence-level QE shared task show that the proposed unified neural network approach consistently outperforms the predictor and estimator approach and significantly outperforms the other baseline QE approaches.