Fair NLP Models with Differentially Private Text Encoders

Gaurav Maheshwari, Pascal Denis, Mikaela Keller, Aurélien Bellet


Abstract
Encoded text representations often capture sensitive attributes about individuals (e.g., race or gender), which raise privacy concerns and can make downstream models unfair to certain groups. In this work, we propose FEDERATE, an approach that combines ideas from differential privacy and adversarial training to learn private text representations which also induces fairer models. We empirically evaluate the trade-off between the privacy of the representations and the fairness and accuracy of the downstream model on four NLP datasets. Our results show that FEDERATE consistently improves upon previous methods, and thus suggest that privacy and fairness can positively reinforce each other.
Anthology ID:
2022.findings-emnlp.514
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2022
Month:
December
Year:
2022
Address:
Abu Dhabi, United Arab Emirates
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6913–6930
Language:
URL:
https://aclanthology.org/2022.findings-emnlp.514
DOI:
Bibkey:
Cite (ACL):
Gaurav Maheshwari, Pascal Denis, Mikaela Keller, and Aurélien Bellet. 2022. Fair NLP Models with Differentially Private Text Encoders. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 6913–6930, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.
Cite (Informal):
Fair NLP Models with Differentially Private Text Encoders (Maheshwari et al., Findings 2022)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2022.findings-emnlp.514.pdf