Zero-Resource Translation with Multi-Lingual Neural Machine Translation

2016/06/13 by Orhan Fırat, Firat, Orhan, Baskaran Sankaran +7 · 6 citations
Computer Science · #Natural Language Processing Techniques #Topic Modeling #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.1606.04164

Abstract

In this paper, we propose a novel finetuning algorithm for the recently introduced multi-way, mulitlingual neural machine translate that enables zero-resource machine translation. When used together with novel many-to-one translation strategies, we empirically show that this finetuning algorithm allows the multi-way, multilingual model to translate a zero-resource language pair (1) as well as a single-pair neural translation model trained with up to 1M direct parallel sentences of the same language pair and (2) better than pivot-based translation strategy, while keeping only one additional copy of attention-related parameters.

Citations

• Sequence to Sequence Learning with Neural Networks

Cited by

• Sheaf-Laplacian Obstruction and Projection Hardness for Cross-Modal Compatibility on a Modality-Independent Site
• Unsupervised Machine Translation Using Monolingual Corpora Only
• A Study of Multilingual Neural Machine Translation
• Multilingual Neural Machine Translation for Zero-Resource Languages
• Google's Multilingual Neural Machine Translation System: Enabling\n Zero-Shot Translation
• A Survey on Low-Resource Neural Machine Translation

Related

• Multi-Way, Multilingual Neural Machine Translation with a Shared Attention Mechanism
• A Study of Multilingual Neural Machine Translation
• Effective Strategies in Zero-Shot Neural Machine Translation
• Multilingual Neural Machine Translation for Zero-Resource Languages
• Google's Multilingual Neural Machine Translation System: Enabling\n Zero-Shot Translation