UDALM: Unsupervised Domain Adaptation through Language Modeling

Constantinos Karouzos, Georgios Paraskevopoulos, Alexandros Potamianos


Abstract
In this work we explore Unsupervised Domain Adaptation (UDA) of pretrained language models for downstream tasks. We introduce UDALM, a fine-tuning procedure, using a mixed classification and Masked Language Model loss, that can adapt to the target domain distribution in a robust and sample efficient manner. Our experiments show that performance of models trained with the mixed loss scales with the amount of available target data and the mixed loss can be effectively used as a stopping criterion during UDA training. Furthermore, we discuss the relationship between A-distance and the target error and explore some limitations of the Domain Adversarial Training approach. Our method is evaluated on twelve domain pairs of the Amazon Reviews Sentiment dataset, yielding 91.74% accuracy, which is an 1.11% absolute improvement over the state-of-the-art.
Anthology ID:
2021.naacl-main.203
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
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2579–2590
Language:
URL:
https://aclanthology.org/2021.naacl-main.203
DOI:
10.18653/v1/2021.naacl-main.203
Bibkey:
Cite (ACL):
Constantinos Karouzos, Georgios Paraskevopoulos, and Alexandros Potamianos. 2021. UDALM: Unsupervised Domain Adaptation through Language Modeling. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2579–2590, Online. Association for Computational Linguistics.
Cite (Informal):
UDALM: Unsupervised Domain Adaptation through Language Modeling (Karouzos et al., NAACL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.naacl-main.203.pdf
Video:
 https://preview.aclanthology.org/ingestion-script-update/2021.naacl-main.203.mp4
Code
 ckarouzos/slp_daptmlm
Data
Multi-Domain Sentiment Dataset v2.0