Domain-adversarial Transfer Learning for Low-resource Neural Machine Translation
When the out-of-domain and in-domain represent different languages,the differences between languages will make it difficult adapt the out-of-domain knowledge to the in-domain.This paper proposes a domain-adversarial transfer learning method to improve the neural machine translation model.Under the adversarial learning frame-work,a domain discriminator is employed to predict the semantic features that from out-of-domain or in-domain,and the encoder is optimized by minimizing the prediction values of the semantic features.When the predicted values of semantic features in the two domains are similar,it means that the model has learned the mapping function that can transfer in-domain data into out-of-domain.Experiments show a certain generalization ability of this method on Mongolian-Chinese and Uyghur-Chinese translation tasks.
万方数据
维普
