Unsupervised Neural Machine Translation for Low-Resource Domains via Meta-Learning

TL;DR

This paper introduces a meta-learning approach for unsupervised neural machine translation that enhances adaptation to low-resource domains using minimal data, outperforming existing transfer methods.

Contribution

It proposes a novel meta-learning algorithm that leverages high-resource domain knowledge to improve low-resource unsupervised translation performance.

Findings

Outperforms transfer learning by 2-4 BLEU scores.

Demonstrates effective fast adaptation to low-resource domains.

Consistently outperforms baseline models in experiments.

Abstract

Unsupervised machine translation, which utilizes unpaired monolingual corpora as training data, has achieved comparable performance against supervised machine translation. However, it still suffers from data-scarce domains. To address this issue, this paper presents a novel meta-learning algorithm for unsupervised neural machine translation (UNMT) that trains the model to adapt to another domain by utilizing only a small amount of training data. We assume that domain-general knowledge is a significant factor in handling data-scarce domains. Hence, we extend the meta-learning algorithm, which utilizes knowledge learned from high-resource domains, to boost the performance of low-resource UNMT. Our model surpasses a transfer learning-based approach by up to 2-4 BLEU scores. Extensive experimental results show that our proposed algorithm is pertinent for fast adaptation and consistently outperforms other baseline models.