徐聪等：文本生成领域的深度强化学习研究进展 411 [79]Xiong W,Hoang T,Wang W Y.DeepPath:a reinforcement [81]Feng J,Huang M L,Zhao L,et al.Reinforcement learning for learning method for knowledge graph reasoning /Proceedings of relation classification from noisy data /Proceedings of Thirty- the 2017 Conference on Empirical Methods in Natural Language Second AAAl Conference on Artificial Intelligence.New Orleans, Processing.Copenhagen,2017:564 2018:5779 [80]Buck C,Bulian J,Ciaramita M,et al.Ask the right questions: [82]Zhang T Y,Huang M L.Zhao L.Learning structured active question reformulation with reinforcement learning.arXiv representation for text classification via reinforcement learning// Preprint (2018-03-02)[2019-06-16].https://arxiv.org/abs/1705. Proceedings of Thirty-Second AAAl Conference on Artificial 07830 Intelligence.New Orleans,2018:6053Xiong  W,  Hoang  T,  Wang  W  Y.  DeepPath:  a  reinforcement learning method for knowledge graph reasoning // Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing. Copenhagen, 2017: 564 [79] Buck  C,  Bulian  J,  Ciaramita  M,  et  al.  Ask  the  right  questions: active  question  reformulation  with  reinforcement  learning. arXiv Preprint (2018-03-02)  [2019-06-16]. https://arxiv.org/abs/1705. 07830 [80] Feng  J,  Huang  M  L,  Zhao  L,  et  al.  Reinforcement  learning  for relation  classification  from  noisy  data  // Proceedings of Thirty￾Second AAAI Conference on Artificial Intelligence. New Orleans, 2018: 5779 [81] Zhang  T  Y,  Huang  M  L,  Zhao  L.  Learning  structured representation  for  text  classification via reinforcement  learning  // Proceedings of Thirty-Second AAAI Conference on Artificial Intelligence. New Orleans, 2018: 6053 [82] 徐    聪等： 文本生成领域的深度强化学习研究进展 · 411 ·