Deep Reinforcement Learning for Chinese Zero Pronoun Resolution
Deep Reinforcement Learning for Chinese Zero Pronoun Resolution
复制标题
DOI：
10.18653/v1/p18-1053
复制
发表时间：
2018-06
期刊：
影响因子：
--
通讯作者：
Qingyu Yin;Yu Zhang;Weinan Zhang;Ting Liu;William Yang Wang
中科院分区：
文献类型：
--
作者：
Qingyu Yin;Yu Zhang;Weinan Zhang;Ting Liu;William Yang Wang
最近的中文零代词解析神经网络模型通过捕获零代词和候选先行词的语义信息获得了很好的性能，但往往是短视的，仅通过做出局部决策来操作。他们通常一次预测零代词和一个候选先行词之间的共指联系，而忽略它们对未来决策的影响。理想情况下，对前面潜在先行词的有用信息进行建模对于对后面的零代词候选先行词对进行分类至关重要，这种需求导致传统的零代词解析模型利用强化学习。在本文中，我们展示了如何整合这些目标，应用深度强化学习来处理任务。在强化学习代理的帮助下，我们的系统学习以顺序方式选择前因的策略，其中早期预测的前因提供的有用信息可用于做出后续的共指决策。 OntoNotes 5.0 上的实验结果表明，我们的方法在三种实验设置下明显优于最先进的方法。
Recent neural network models for Chinese zero pronoun resolution gain great performance by capturing semantic information for zero pronouns and candidate antecedents, but tend to be short-sighted, operating solely by making local decisions. They typically predict coreference links between the zero pronoun and one single candidate antecedent at a time while ignoring their influence on future decisions. Ideally, modeling useful information of preceding potential antecedents is crucial for classifying later zero pronoun-candidate antecedent pairs, a need which leads traditional models of zero pronoun resolution to draw on reinforcement learning. In this paper, we show how to integrate these goals, applying deep reinforcement learning to deal with the task. With the help of the reinforcement learning agent, our system learns the policy of selecting antecedents in a sequential manner, where useful information provided by earlier predicted antecedents could be utilized for making later coreference decisions. Experimental results on OntoNotes 5.0 show that our approach substantially outperforms the state-of-the-art methods under three experimental settings.