Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLP

Trapit Bansal, Karthick Prasad Gunasekaran, Tong Wang, Tsendsuren Munkhdalai, Andrew McCallum


Abstract
Meta-learning considers the problem of learning an efficient learning process that can leverage its past experience to accurately solve new tasks. However, the efficacy of meta-learning crucially depends on the distribution of tasks available for training, and this is often assumed to be known a priori or constructed from limited supervised datasets. In this work, we aim to provide task distributions for meta-learning by considering self-supervised tasks automatically proposed from unlabeled text, to enable large-scale meta-learning in NLP. We design multiple distributions of self-supervised tasks by considering important aspects of task diversity, difficulty, type, domain, and curriculum, and investigate how they affect meta-learning performance. Our analysis shows that all these factors meaningfully alter the task distribution, some inducing significant improvements in downstream few-shot accuracy of the meta-learned models. Empirically, results on 20 downstream tasks show significant improvements in few-shot learning – adding up to +4.2% absolute accuracy (on average) to the previous unsupervised meta-learning method, and perform comparably to supervised methods on the FewRel 2.0 benchmark.
Anthology ID:
2021.emnlp-main.469
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5812–5824
Language:
URL:
https://aclanthology.org/2021.emnlp-main.469
DOI:
10.18653/v1/2021.emnlp-main.469
Bibkey:
Cite (ACL):
Trapit Bansal, Karthick Prasad Gunasekaran, Tong Wang, Tsendsuren Munkhdalai, and Andrew McCallum. 2021. Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLP. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5812–5824, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLP (Bansal et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.469.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.469.mp4
Data
CCNetFewRelFewRel 2.0GLUE