Adversarial Neural Machine Translation
Lijun Wu1, Yingce Xia2, Li Zhao3, Fei Tian3, Tao Qin3, Jianhuang Lai1,4 Tie-Yan Liu3 1School of Data and Computer Science, Sun Yat-sen University 2University of Science and Technology of China 3Microsoft Research Asia 4Guangdong Key Laboratory of Information Security Technology
Abstract
In this paper, we study a new learning paradigm for Neural Machine Translation (NMT). Instead of maximizing the likelihood of the human translation as in previous works, we minimize the distinction between human translation and the translation given by an NMT model. To achieve this goal, inspired by the recent success of Generative Adversarial Networks (GANs), we employ an adversarial training architecture and name it as Adversarial-NMT. In Adversarial-NMT, the training of the NMT model is assisted by an adversary, which is an elaborately designed Convolutional Neural Network (CNN). The goal of the adversary is to differentiate the translation result generated by the NMT model from that by human. The goal of the NMT model is to produce high quality translations so as to cheat the adversary. A policy gradient method is leveraged to co-train the NMT model and the adversary. Experimental results on English $\rightarrow$ French and German $\rightarrow$ English translation tasks show that Adversarial-NMT can achieve significantly better translation quality than several strong baselines.
中文速览
神经机器翻译（NMT）长期依赖最大似然估计训练，导致机器译文与人工译文之间仍存在明显差距。这篇论文借鉴生成对抗网络（GAN）的思路，提出了 Adversarial-NMT：让一个专门设计的卷积神经网络（CNN）充当"鉴别器"，负责区分机器译文和人工译文；而 NMT 模型则作为"生成器"，不断改进翻译质量以欺骗鉴别器；由于译文是离散序列无法直接反向传播，作者引入强化学习中的策略梯度方法（REINFORCE）来联合训练两个模块。在英法和德英翻译任务上的实验表明，该方法显著优于传统 NMT 及基于 BLEU 直接优化的多个强基线。这项工作的意义在于，它首次系统地将对抗训练引入序列到序列的机器翻译任务，为摆脱 BLEU 等固定评价指标的束缚、让机器译文向人工译文质量靠拢提供了一条新路径。
原文 arXiv:1704.06933;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1704.06933v4