Debiasing Neural Retrieval via In-batch Balancing Regularization

Yuantong Li, Xiaokai Wei, Zijian Wang, Shen Wang, Parminder Bhatia, Xiaofei Ma, Andrew Arnold


Abstract
People frequently interact with information retrieval (IR) systems, however, IR models exhibit biases and discrimination towards various demographics. The in-processing fair ranking methods provides a trade-offs between accuracy and fairness through adding a fairness-related regularization term in the loss function. However, there haven’t been intuitive objective functions that depend on the click probability and user engagement to directly optimize towards this. In this work, we propose the In-Batch Balancing Regularization (IBBR) to mitigate the ranking disparity among subgroups. In particular, we develop a differentiable normed Pairwise Ranking Fairness (nPRF) and leverage the T-statistics on top of nPRF over subgroups as a regularization to improve fairness. Empirical results with the BERT-based neural rankers on the MS MARCO Passage Retrieval dataset with the human-annotated non-gendered queries benchmark (CITATION) show that our IBBR method with nPRF achieves significantly less bias with minimal degradation in ranking performance compared with the baseline.
Anthology ID:
2022.gebnlp-1.5
Volume:
Proceedings of the 4th Workshop on Gender Bias in Natural Language Processing (GeBNLP)
Month:
July
Year:
2022
Address:
Seattle, Washington
Editors:
Christian Hardmeier, Christine Basta, Marta R. Costa-jussà, Gabriel Stanovsky, Hila Gonen
Venue:
GeBNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
58–66
Language:
URL:
https://aclanthology.org/2022.gebnlp-1.5
DOI:
10.18653/v1/2022.gebnlp-1.5
Bibkey:
Cite (ACL):
Yuantong Li, Xiaokai Wei, Zijian Wang, Shen Wang, Parminder Bhatia, Xiaofei Ma, and Andrew Arnold. 2022. Debiasing Neural Retrieval via In-batch Balancing Regularization. In Proceedings of the 4th Workshop on Gender Bias in Natural Language Processing (GeBNLP), pages 58–66, Seattle, Washington. Association for Computational Linguistics.
Cite (Informal):
Debiasing Neural Retrieval via In-batch Balancing Regularization (Li et al., GeBNLP 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.gebnlp-1.5.pdf
Data
MS MARCO