Zhuang Li

Zhuang Li

Zhuang is a Research Fellow at Monash University.

Unsupervised Pre-training With Seq2Seq Reconstruction Loss for Deep Relation Extraction Models

Published in ALTA, 2016

Recommended citation: Li, Z., Qu, L., Xu, Q., & Johnson, M. Unsupervised Pre-training with Sequence Reconstruction Loss for Deep Relation Extraction Models. Workshop of The Australasian Language Technology Association. https://www.aclweb.org/anthology/U16-1006/

Relation extraction models based on deep learning have been attracting a lot of attention recently. Little research is carried out to reduce their need of labeled training data. In this work, we propose an unsupervised pre-training method based on the sequence-to-sequence model for deep relation extraction models. The pre-trained models need only half or even less training data to achieve equivalent performance as the same models without pre-training.

Download paper here

Citation:

@inproceedings{li-etal-2016-unsupervised,
    title = "Unsupervised Pre-training With {S}eq2{S}eq Reconstruction Loss for Deep Relation Extraction Models",
    author = "Li, Zhuang  and
      Qu, Lizhen  and
      Xu, Qiongkai  and
      Johnson, Mark",
    booktitle = "Proceedings of the Australasian Language Technology Association Workshop 2016",
    month = dec,
    year = "2016",
    address = "Melbourne, Australia",
    url = "https://www.aclweb.org/anthology/U16-1006",
    pages = "54--64",
}