2018/11/06 by Hokamp, Chris, Ruder, Sebastian, Glover, John
#Computation and Language (cs.CL) #FOS: Computer and information sciences
paper · doi:10.48550/arxiv.1811.02278
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.