Unsupervised Neural Machine Translation Initialized by Unsupervised Statistical Machine Translation
This work addresses the problem of machine translation without parallel data, offering an incremental improvement for researchers and practitioners in NLP.
The authors tackled unsupervised neural machine translation by initializing training with synthetic bilingual data from unsupervised statistical machine translation and improving it with back-translation, achieving a new state-of-the-art on the WMT16 German-English task.
Recent work achieved remarkable results in training neural machine translation (NMT) systems in a fully unsupervised way, with new and dedicated architectures that rely on monolingual corpora only. In this work, we propose to define unsupervised NMT (UNMT) as NMT trained with the supervision of synthetic bilingual data. Our approach straightforwardly enables the use of state-of-the-art architectures proposed for supervised NMT by replacing human-made bilingual data with synthetic bilingual data for training. We propose to initialize the training of UNMT with synthetic bilingual data generated by unsupervised statistical machine translation (USMT). The UNMT system is then incrementally improved using back-translation. Our preliminary experiments show that our approach achieves a new state-of-the-art for unsupervised machine translation on the WMT16 German--English news translation task, for both translation directions.