Evaluating the Values of Sources in Transfer Learning

Md Rizwan Parvez, Kai-Wei Chang


Abstract
Transfer learning that adapts a model trained on data-rich sources to low-resource targets has been widely applied in natural language processing (NLP). However, when training a transfer model over multiple sources, not every source is equally useful for the target. To better transfer a model, it is essential to understand the values of the sources. In this paper, we develop , an efficient source valuation framework for quantifying the usefulness of the sources (e.g., ) in transfer learning based on the Shapley value method. Experiments and comprehensive analyses on both cross-domain and cross-lingual transfers demonstrate that our framework is not only effective in choosing useful transfer sources but also the source values match the intuitive source-target similarity.
Anthology ID:
2021.naacl-main.402
Volume:
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
June
Year:
2021
Address:
Online
Editors:
Kristina Toutanova, Anna Rumshisky, Luke Zettlemoyer, Dilek Hakkani-Tur, Iz Beltagy, Steven Bethard, Ryan Cotterell, Tanmoy Chakraborty, Yichao Zhou
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5084–5116
Language:
URL:
https://aclanthology.org/2021.naacl-main.402
DOI:
10.18653/v1/2021.naacl-main.402
Bibkey:
Cite (ACL):
Md Rizwan Parvez and Kai-Wei Chang. 2021. Evaluating the Values of Sources in Transfer Learning. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5084–5116, Online. Association for Computational Linguistics.
Cite (Informal):
Evaluating the Values of Sources in Transfer Learning (Parvez & Chang, NAACL 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.naacl-main.402.pdf
Video:
 https://aclanthology.org/2021.naacl-main.402.mp4
Code
 rizwan09/NLPDV
Data
GLUEMulti-Domain SentimentMultiNLIQNLIUniversal DependenciesXNLI