Abstract
Task variance regularization, which can be used to improve the generalization of Multi-task Learning (MTL) models, remains unexplored in multi-task text classification. Accordingly, to fill this gap, this paper investigates how the task might be effectively regularized, and consequently proposes a multi-task learning method based on adversarial multi-armed bandit. The proposed method, named BanditMTL, regularizes the task variance by means of a mirror gradient ascent-descent algorithm. Adopting BanditMTL in the multi-task text classification context is found to achieve state-of-the-art performance. The results of extensive experiments back up our theoretical analysis and validate the superiority of our proposals.- Anthology ID:
- 2021.acl-long.428
- Volume:
- Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
- Month:
- August
- Year:
- 2021
- Address:
- Online
- Editors:
- Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
- Venues:
- ACL | IJCNLP
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5506–5516
- Language:
- URL:
- https://aclanthology.org/2021.acl-long.428
- DOI:
- 10.18653/v1/2021.acl-long.428
- Cite (ACL):
- Yuren Mao, Zekai Wang, Weiwei Liu, Xuemin Lin, and Wenbin Hu. 2021. BanditMTL: Bandit-based Multi-task Learning for Text Classification. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 5506–5516, Online. Association for Computational Linguistics.
- Cite (Informal):
- BanditMTL: Bandit-based Multi-task Learning for Text Classification (Mao et al., ACL-IJCNLP 2021)
- PDF:
- https://preview.aclanthology.org/emnlp22-frontmatter/2021.acl-long.428.pdf