Trivial Transfer Learning for Low-Resource Neural Machine Translation
arXiv:1809.00357 · doi:10.18653/v1/W18-6325
Abstract
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 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.
Accepted to WMT18 reseach paper, Proceedings of the 3rd Conference on Machine Translation 2018
Cited by in corpus (4)
- Non-Fluent Synthetic Target-Language Data Improve Neural Machine Translation
- Tag-less Back-Translation
- Cross-Sign Language Transfer Learning Using Domain Adaptation with Multi-scale Temporal Alignment
- Quantity vs. Quality of Monolingual Source Data in Automatic Text Translation: Can It Be Too Little If It Is Too Good?