Abstract
Multi-task learning and self-training are two common ways to improve a machine learning model’s performance in settings with limited training data. Drawing heavily on ideas from those two approaches, we suggest transductive auxiliary task self-training: training a multi-task model on (i) a combination of main and auxiliary task training data, and (ii) test instances with auxiliary task labels which a single-task version of the model has previously generated. We perform extensive experiments on 86 combinations of languages and tasks. Our results are that, on average, transductive auxiliary task self-training improves absolute accuracy by up to 9.56% over the pure multi-task model for dependency relation tagging and by up to 13.03% for semantic tagging.- Anthology ID:
- D19-6128
- Volume:
- Proceedings of the 2nd Workshop on Deep Learning Approaches for Low-Resource NLP (DeepLo 2019)
- Month:
- November
- Year:
- 2019
- Address:
- Hong Kong, China
- Venue:
- WS
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 253–258
- Language:
- URL:
- https://aclanthology.org/D19-6128
- DOI:
- 10.18653/v1/D19-6128
- Cite (ACL):
- Johannes Bjerva, Katharina Kann, and Isabelle Augenstein. 2019. Transductive Auxiliary Task Self-Training for Neural Multi-Task Models. In Proceedings of the 2nd Workshop on Deep Learning Approaches for Low-Resource NLP (DeepLo 2019), pages 253–258, Hong Kong, China. Association for Computational Linguistics.
- Cite (Informal):
- Transductive Auxiliary Task Self-Training for Neural Multi-Task Models (Bjerva et al., 2019)
- PDF:
- https://preview.aclanthology.org/paclic-22-ingestion/D19-6128.pdf
- Data
- Universal Dependencies