Neural Self-Training through Spaced Repetition

Hadi Amiri


Abstract
Self-training is a semi-supervised learning approach for utilizing unlabeled data to create better learners. The efficacy of self-training algorithms depends on their data sampling techniques. The majority of current sampling techniques are based on predetermined policies which may not effectively explore the data space or improve model generalizability. In this work, we tackle the above challenges by introducing a new data sampling technique based on spaced repetition that dynamically samples informative and diverse unlabeled instances with respect to individual learner and instance characteristics. The proposed model is specifically effective in the context of neural models which can suffer from overfitting and high-variance gradients when trained with small amount of labeled data. Our model outperforms current semi-supervised learning approaches developed for neural networks on publicly-available datasets.
Anthology ID:
N19-1003
Volume:
Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
Month:
June
Year:
2019
Address:
Minneapolis, Minnesota
Editors:
Jill Burstein, Christy Doran, Thamar Solorio
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
21–31
Language:
URL:
https://aclanthology.org/N19-1003
DOI:
10.18653/v1/N19-1003
Bibkey:
Cite (ACL):
Hadi Amiri. 2019. Neural Self-Training through Spaced Repetition. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 21–31, Minneapolis, Minnesota. Association for Computational Linguistics.
Cite (Informal):
Neural Self-Training through Spaced Repetition (Amiri, NAACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/N19-1003.pdf
Video:
 https://aclanthology.org/N19-1003.mp4
Data
IMDb Movie Reviews