Semi-supervised structured prediction with neural CRF autoencoder
Zhang, Xiao1; Jiang, Yong2; Peng, Hao1; Tu, Kewei2; Goldwasser, Dan1
2017
Source Publication2017 CONFERENCE ON EMPIRICAL METHODS IN NATURAL LANGUAGE PROCESSING, EMNLP 2017
Pages1701-1711
Status已发表
AbstractIn this paper we propose an end-to-end neural CRF autoencoder (NCRF-AE) model for semi-supervised learning of sequential structured prediction problems. Our NCRF-AE consists of two parts: an encoder which is a CRF model enhanced by deep neural networks, and a decoder which is a generative model trying to reconstruct the input. Our model has a unified structure with different loss functions for labeled and unlabeled data with shared parameters. We developed a variation of the EM algorithm for optimizing both the encoder and the decoder simultaneously by decoupling their parameters. Our experimental results over the Part-of-Speech (POS) tagging task on eight different languages, show that the NCRF-AE model can outperform competitive systems in both supervised and semi-supervised scenarios.
© 2017 Association for Computational Linguistics.
Conference PlaceCopenhagen, Denmark
Indexed ByEI
PublisherAssociation for Computational Linguistics (ACL)
EI Accession Number20194207538747
EI KeywordsDecoding ; Deep neural networks ; Learning algorithms ; Machine learning ; Signal encoding ; Supervised learning
EI Classification NumberInformation Theory and Signal Processing:716.1 ; Data Processing and Image Processing:723.2
Original Document TypeConference article (CA)
Document Type会议论文
Identifierhttps://kms.shanghaitech.edu.cn/handle/2MSLDSTB/29393
Collection信息科学与技术学院_博士生
信息科学与技术学院_PI研究组_屠可伟组
Affiliation1.Department of Computer Science, Purdue University, West Lafayette, United States
2.School of Information Science and Technology, ShanghaiTech University, Shanghai, China
Recommended Citation
GB/T 7714
Zhang, Xiao,Jiang, Yong,Peng, Hao,et al. Semi-supervised structured prediction with neural CRF autoencoder[C]:Association for Computational Linguistics (ACL),2017:1701-1711.
Files in This Item: Download All
File Name/Size DocType Version Access License
10.18653@v1@D17-1179(571KB)会议论文出版稿开放获取UnknownView Download
Related Services
Usage statistics
Scholar Google
Similar articles in Scholar Google
[Zhang, Xiao]'s Articles
[Jiang, Yong]'s Articles
[Peng, Hao]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhang, Xiao]'s Articles
[Jiang, Yong]'s Articles
[Peng, Hao]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhang, Xiao]'s Articles
[Jiang, Yong]'s Articles
[Peng, Hao]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 10.18653@v1@D17-1179.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.