OSCaR: Orthogonal Subspace Correction and Rectification of Biases in Word Embeddings

Sunipa Dev, Tao Li, Jeff M Phillips, Vivek Srikumar


Abstract
Language representations are known to carry stereotypical biases and, as a result, lead to biased predictions in downstream tasks. While existing methods are effective at mitigating biases by linear projection, such methods are too aggressive: they not only remove bias, but also erase valuable information from word embeddings. We develop new measures for evaluating specific information retention that demonstrate the tradeoff between bias removal and information retention. To address this challenge, we propose OSCaR (Orthogonal Subspace Correction and Rectification), a bias-mitigating method that focuses on disentangling biased associations between concepts instead of removing concepts wholesale. Our experiments on gender biases show that OSCaR is a well-balanced approach that ensures that semantic information is retained in the embeddings and bias is also effectively mitigated.
Anthology ID:
2021.emnlp-main.411
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
5034–5050
Language:
URL:
https://aclanthology.org/2021.emnlp-main.411
DOI:
10.18653/v1/2021.emnlp-main.411
Bibkey:
Cite (ACL):
Sunipa Dev, Tao Li, Jeff M Phillips, and Vivek Srikumar. 2021. OSCaR: Orthogonal Subspace Correction and Rectification of Biases in Word Embeddings. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 5034–5050, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
OSCaR: Orthogonal Subspace Correction and Rectification of Biases in Word Embeddings (Dev et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.411.pdf
Code
 Shaul1321/nullspace_projection
Data
MultiNLISNLI