DualFair: Fair Representation Learning at Both Group and Individual Levels via Contrastive Self-supervision

TL;DR

DualFair is a self-supervised learning model that simultaneously achieves group and individual fairness in representations, improving fairness in Web applications by using contrastive loss and self-knowledge distillation.

Contribution

It introduces a novel approach that jointly optimizes for group and counterfactual fairness using contrastive self-supervision, which is a significant advancement over existing single-fairness models.

Findings

Effective debiasing of sensitive attributes like gender and race.

Joint fairness optimization improves overall fairness metrics.

Model maintains high representation quality for downstream tasks.

Abstract

Algorithmic fairness has become an important machine learning problem, especially for mission-critical Web applications. This work presents a self-supervised model, called DualFair, that can debias sensitive attributes like gender and race from learned representations. Unlike existing models that target a single type of fairness, our model jointly optimizes for two fairness criteria - group fairness and counterfactual fairness - and hence makes fairer predictions at both the group and individual levels. Our model uses contrastive loss to generate embeddings that are indistinguishable for each protected group, while forcing the embeddings of counterfactual pairs to be similar. It then uses a self-knowledge distillation method to maintain the quality of representation for the downstream tasks. Extensive analysis over multiple datasets confirms the model's validity and further shows the synergy of jointly addressing two fairness criteria, suggesting the model's potential value in fair intelligent Web applications.