Fairness-aware Class Imbalanced Learning
This work addresses fairness and class imbalance in NLP, but it is incremental as it builds on existing long-tail learning methods.
The paper tackles the problem of class imbalance and demographic bias in NLP tasks by extending a margin-loss based approach with fairness-enforcing methods, showing through experiments on tweet sentiment and occupation classification that the proposed methods help mitigate both issues.
Class imbalance is a common challenge in many NLP tasks, and has clear connections to bias, in that bias in training data often leads to higher accuracy for majority groups at the expense of minority groups. However there has traditionally been a disconnect between research on class-imbalanced learning and mitigating bias, and only recently have the two been looked at through a common lens. In this work we evaluate long-tail learning methods for tweet sentiment and occupation classification, and extend a margin-loss based approach with methods to enforce fairness. We empirically show through controlled experiments that the proposed approaches help mitigate both class imbalance and demographic biases.