In neural machine translation (NMT), the most common practice is to stack a number of recurrent or feed-forward layers in the encoder and the decoder. As a result, the addition of each new layer improves the translation quality significantly. However, this also leads to a significant increase in the number of parameters. In this paper, we propose to share parameters across all the layers thereby leading to a recurrently stacked NMT model. We empirically show that the translation quality of a model that recurrently stacks a single layer 6 times is comparable to the translation quality of a model that stacks 6 separate layers. We also show that using pseudo-parallel corpora by back-translation leads to further significant improvements in translation quality.
@article{arxiv.1807.05353,
title = {Recurrent Stacking of Layers for Compact Neural Machine Translation Models},
author = {Raj Dabre and Atsushi Fujita},
journal= {arXiv preprint arXiv:1807.05353},
year = {2018}
}
Comments
Version 2 (Current): Fixed Typos. Additional Results for models using back-translated data. Resized the figure. Better explanations of some parts. Version 1: Initial version