Mitigating Gender Bias Amplification in Distribution by Posterior Regularization

TL;DR

This paper investigates gender bias amplification in NLP models from a distribution perspective and proposes a posterior regularization method to mitigate this bias with minimal performance loss.

Contribution

It introduces a novel distribution-based analysis of bias amplification and presents a posterior regularization technique to effectively reduce bias in model predictions.

Findings

Bias is amplified in the predicted probability distribution.

The proposed method nearly removes bias amplification.

Minimal performance loss with effective bias mitigation.

Abstract

Advanced machine learning techniques have boosted the performance of natural language processing. Nevertheless, recent studies, e.g., Zhao et al. (2017) show that these techniques inadvertently capture the societal bias hidden in the corpus and further amplify it. However, their analysis is conducted only on models' top predictions. In this paper, we investigate the gender bias amplification issue from the distribution perspective and demonstrate that the bias is amplified in the view of predicted probability distribution over labels. We further propose a bias mitigation approach based on posterior regularization. With little performance loss, our method can almost remove the bias amplification in the distribution. Our study sheds the light on understanding the bias amplification.