arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1808.08013cs.IRcs.LGstat.ML
收藏

基于强化学习的噪声数据关系分类

Reinforcement Learning for Relation Classification from Noisy Data

Microsoft Research Asia(微软亚洲研究院);College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院);Tsinghua University(清华大学)

Jun Feng, Minlie Huang, Li Zhao, Yang Yang, Xiaoyan Zhu

摘要:

现有的依赖于远程监督的关系分类方法假设一袋提到实体对的句子都描述了实体对的关系。这样的方法,在袋级执行分类,不能识别关系和句子之间的映射,并在很大程度上遭受噪声标记问题。在本文中,我们提出了一种新的模型,用于从有噪数据中进行句子级别的关系分类。该模型有两个模块:实例选择器和关系分类器。实例选择器通过强化学习选择出高质量的句子,并将选择出的句子输入到关系分类器中,关系分类器进行句子级别预测并向实例选择器提供奖励。这两个模块联合训练,以优化实例选择和关系分类过程。实验结果表明,该模型能有效地处理数据中的噪声,在句子层次上取得较好的关系分类性能。

英文摘要:

Existing relation classification methods that rely on distant supervision assume that a bag of sentences mentioning an entity pair are all describing a relation for the entity pair. Such methods, performing classification at the bag level, cannot identify the mapping between a relation and a sentence, and largely suffers from the noisy labeling problem. In this paper, we propose a novel model for relation classification at the sentence level from noisy data. The model has two modules: an instance selector and a relation classifier. The instance selector chooses high-quality sentences with reinforcement learning and feeds the selected sentences into the relation classifier, and the relation classifier makes sentence level prediction and provides rewards to the instance selector. The two modules are trained jointly to optimize the instance selection and relation classification processes. Experiment results show that our model can deal with the noise of data effectively and obtains better performance for relation classification at the sentence level.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑