papers AI Learner
The Github is limit! Click to go to the new site.

Target Conditioned Sampling: Optimizing Data Selection for Multilingual Neural Machine Translation

2019-05-20
Xinyi Wang, Graham Neubig

Abstract

To improve low-resource Neural Machine Translation (NMT) with multilingual corpora, training on the most related high-resource language only is often more effective than using all data available (Neubig and Hu, 2018). However, it is possible that an intelligent data selection strategy can further improve low-resource NMT with data from other auxiliary languages. In this paper, we seek to construct a sampling distribution over all multilingual data, so that it minimizes the training loss of the low-resource language. Based on this formulation, we propose an efficient algorithm, Target Conditioned Sampling (TCS), which first samples a target sentence, and then conditionally samples its source sentence. Experiments show that TCS brings significant gains of up to 2 BLEU on three of four languages we test, with minimal training overhead.

Abstract (translated by Google)
URL

http://arxiv.org/abs/1905.08212

PDF

http://arxiv.org/pdf/1905.08212


Comments

Content