We frame unsupervised machine translation (MT) in the context of multi-task learning (MTL), combining insights from both directions. We leverage off-the-shelf neural MT architectures to train unsupervised MT models with no parallel data and show that such models can achieve reasonably good performance, competitive with models purpose-built for unsupervised MT. Finally, we propose improvements that allow us to apply our models to English-Turkish, a truly low-resource language pair.
@article{arxiv.1811.02278,
title = {Off-the-Shelf Unsupervised NMT},
author = {Chris Hokamp and Sebastian Ruder and John Glover},
journal= {arXiv preprint arXiv:1811.02278},
year = {2018}
}