header

Curriculum Learning for Distant Supervision Relation Extraction

10 Pages Posted: 29 Sep 2020 Publication Status: Accepted

See all articles by Liu Qiongxin

Liu Qiongxin

Beijing Institute of Technology - Beijing Engineering Research Center of High Volume Lanuage Information Processing and Cloud Computing Applications

Wang Peng

School of Computer Science and Technology, Beijing Institute of Technology

Wang Jiasheng

School of Computer Science and Technology, Beijing Institute of Technology

Ma Jing

School of Computer Science and Technology, Beijing Institute of Technology

Abstract

Relation extraction under distant supervision leverages the existing knowledge base to label data automatically, thus greatly reduced the consumption of human labors. Although distant supervision is an efficient method, to obtain a large amount of labeled data, the training dataset labeled by distant supervision suffers from noise problem resulting in poor generalization ability of the relation extractor. To alleviate the noise problem, we propose a novel relation extraction method based on curriculum learning. Curriculum learning is utilized to guide the training process of relation extractor, specifically through the predefined curriculum-driven mentor network. Mentor network can dynamically adjust the weights of sentences during training, giving lower weights to noisy sentences and higher eights to truly labeled sentences. Relation extractor and mentor network are trained collaboratively to optimize joint objective. The experimental results show that the proposed method can improve the generalization ability of relation extractor in a noisy environment and obtains better performance for relation extraction.

Keywords: relation extraction, curriculum learning, mentor network

Suggested Citation

Qiongxin, Liu and Peng, Wang and Jiasheng, Wang and Jing, Ma, Curriculum Learning for Distant Supervision Relation Extraction (September 22, 2020). Available at SSRN: https://ssrn.com/abstract=3697476 or http://dx.doi.org/10.2139/ssrn.3697476

Liu Qiongxin (Contact Author)

Beijing Institute of Technology - Beijing Engineering Research Center of High Volume Lanuage Information Processing and Cloud Computing Applications ( email )

China

Wang Peng

School of Computer Science and Technology, Beijing Institute of Technology

Wang Jiasheng

School of Computer Science and Technology, Beijing Institute of Technology

Ma Jing

School of Computer Science and Technology, Beijing Institute of Technology

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Abstract Views
86
Downloads
13
PlumX Metrics