Trivial Transfer Learning for Low-Resource Neural Machine Translation

Tom Kocmi, Ondřej Bojar
2018 Proceedings of the Third Conference on Machine Translation: Research Papers  
Transfer learning has been proven as an effective technique for neural machine translation under low-resource conditions. Existing methods require a common target language, language relatedness, or specific training tricks and regimes. We present a simple transfer learning method, where we first train a "parent" model for a high-resource language pair and then continue the training on a lowresource pair only by replacing the training corpus. This "child" model performs significantly better than
more » ... the baseline trained for lowresource pair only. We are the first to show this for targeting different languages, and we observe the improvements even for unrelated languages with different alphabets.
doi:10.18653/v1/w18-6325 dblp:conf/wmt/KocmiB18 fatcat:htdtqruozfddlo7dqrsypzcfqy