Dual Learning for Machine Translation

By Di He, Yingce Xia, Tao Qin, Liwei Wang, Nenghai Yu, Tie-Yan Liu, Wei-Ying Ma. 2016

900 citations recorded by the source corpus.

While neural machine translation (NMT) is making good progress in the past two years, tens of millions of bilingual sentence pairs are needed for its training. However, human labeling is very costly. To tackle this training data bottleneck, we develop a dual-learning mechanism, which can enable an NMT system to automatically learn from unlabeled data through a dual-learning game. This mechanism is inspired by the following observation: any machine translation task has a dual task, e.g., English-to-French translation (primal) versus French-to-English translation (dual); the primal and dual tasks can form a closed loop, and generate informative feedback signals to train the translation models, even if without the involvement of a human labeler. In the dual-learning mechanism, we use one agent to represent the model for the primal task and the other agent to represent the model for the dual task, then ask them to teach each other through a reinforcement learning process. Based on the feedback signals generated during this process (e.g., the language-model likelihood of the output of a model, and the reconstruction error of the original sentence after the primal and dual translations), we can iteratively update the two models until convergence (e.g., using the policy gradient methods). We call the corresponding approach to neural machine translation dual-NMT. Experiments show that dual-NMT works very well on English ↔ French translation; especially, by learning from monolingual data (with 10% bilingual data for warm start), it achieves a comparable accuracy to NMT trained from the full bilingual data for the French-to-English translation task.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Unsupervised Data Augmentation for Consistency Training (2019, background): “Back translation [55, 15] and dual learning [20, 5] can be regarded as performing data augmentation on monolingual data and have been shown to improve the performance for machine translation.”
Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks (2017, methodology): “[55] independently introduce a similar objective for unpaired image-to-image translation, inspired by dual learning in machine translation [15].”
Unsupervised Speech Recognition (2021, background): “, 2019), similar to dual learning in machine translation (Xia et al., 2016).”
Unsupervised Machine Translation Using Monolingual Corpora Only (2017, background): “Perhaps, the most relevant prior work is by He et al. (2016b), who essentially optimizes directly for the model selection metric we propose in section 3.2.”
Unsupervised Neural Machine Translation (2017, methodology): “While NMT systems are typically built for a speciﬁc translation direction (e.g. either French → English or English → French), we exploit the dual nature of machine translation (He et al., 2016; Firat et al., 2016a) and handle both directions together (e.g. French ↔ English).”
Achieving Human Parity on Automatic Chinese to English News Translation (2018, methodology): “Unlike DUL, which aims to effectively leverage monolingual data, DSL is an approach to better utilize bilingual training data by enhancing probabilistic correlations within the two models.”
Phrase-Based & Neural Unsupervised Machine Translation (2018, methodology): “This limited supervision is typically provided as a small set of parallel sentences (Sennrich et al., 2015a; Gul-cehre et al., 2015; He et al., 2016; Gu et al., 2018; Wang et al., 2018); large sets of parallel sentences in related languages (Firat et al., 2016; Johnson et al., 2016; Chen et al.,…”
When and Why Are Pre-Trained Word Embeddings Useful for Neural Machine Translation? (2018, background): “Researchers have worked on a number of meth-ods for using monolingual data in NMT systems (Cheng et al., 2016; He et al., 2016; Ramachan-dran et al., 2016).”
Unsupervised Translation of Programming Languages (2020, background): “Since creating parallel corpora for training is not realistic (creating a small parallel corpus for evaluation is already challenging [19]), some approaches have investigated the use of monolingual data to improve existing machine translation systems [17, 20, 41, 49].”
Deep Mutual Learning (2017, background): “Collaborative Learning Other related ideas on collaborative learning include Dual Learning [6] where two crosslingual translation models teach each other interactively.”
DualGAN: Unsupervised Dual Learning for Image-to-Image Translation (2017, methodology): “[23] to reduce the requirement on labeled data in training Englishto-French and French-to-English translators.”
Ask the Right Questions: Active Question Reformulation with Reinforcement Learning (2017, methodology): “Policy gradient has also been used to define sequence level reward functions based on language models and reconstruction errors to bootstrap MT with less resources [Xia et al., 2016].”