A Study of Reinforcement Learning for Neural Machine Translation

Understand

Recent studies have shown that reinforcement learning (RL) is an effective approach for improving the performance of neural machine translation (NMT) system.

• However, due to its instability, successfully RL training is challenging, especially in real-world systems where deep models and large datasets are leveraged.
• In this paper, taking several large-scale translation tasks as testbeds, we conduct a systematic study on how to train better NMT models using reinforcement learning.
• We provide a comprehensive comparison of several important factors (e.g., baseline reward, reward shaping) in RL training.