FairGAN: Fairness-aware Generative Adversarial Networks

TL;DR

FairGAN is a novel generative adversarial network designed to produce fair, discrimination-free data while maintaining data utility, thereby enabling fair classification on real datasets.

Contribution

Introduces FairGAN, a fairness-aware GAN that generates discrimination-free data and ensures fair classification, advancing fair data synthesis methods.

Findings

FairGAN effectively generates fair data in experiments.

Classifiers trained on FairGAN data achieve fair classification.

FairGAN outperforms naive fair data generation models.

Abstract

Fairness-aware learning is increasingly important in data mining. Discrimination prevention aims to prevent discrimination in the training data before it is used to conduct predictive analysis. In this paper, we focus on fair data generation that ensures the generated data is discrimination free. Inspired by generative adversarial networks (GAN), we present fairness-aware generative adversarial networks, called FairGAN, which are able to learn a generator producing fair data and also preserving good data utility. Compared with the naive fair data generation models, FairGAN further ensures the classifiers which are trained on generated data can achieve fair classification on real data. Experiments on a real dataset show the effectiveness of FairGAN.