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.

本文提出了一种新颖的微调算法，该算法针对最近引入的多方位、多语言神经机器翻译，该翻译使零资源机器翻译成为可能，并且在与新颖的一对多翻译策略结合使用时，我们在经验上表明，该微调算法使得多方位、多语言模型能够翻译一个零资源语言对，且与单一对神经翻译模型训练 1M 条同一语言对的直接平行句子时相当，并且比基于中继的翻译策略更好，同时只保留一个额外的注意到相关参数的副本。