On the Fairness of Generative Adversarial Networks (GANs)

TL;DR

This paper investigates fairness issues in GANs, revealing inherent biases towards certain groups during training and proposing conditioning and ensemble methods to improve group fairness in generated data.

Contribution

It identifies fairness concerns in GANs and introduces novel conditioning and ensemble techniques to mitigate bias and promote equitable data generation.

Findings

GANs may favor certain groups during training

Proposed conditioning and ensemble methods improve fairness

Methods enable more balanced data generation across groups

Abstract

Generative adversarial networks (GANs) are one of the greatest advances in AI in recent years. With their ability to directly learn the probability distribution of data, and then sample synthetic realistic data. Many applications have emerged, using GANs to solve classical problems in machine learning, such as data augmentation, class unbalance problems, and fair representation learning. In this paper, we analyze and highlight fairness concerns of GANs model. In this regard, we show empirically that GANs models may inherently prefer certain groups during the training process and therefore they're not able to homogeneously generate data from different groups during the testing phase. Furthermore, we propose solutions to solve this issue by conditioning the GAN model towards samples' group or using ensemble method (boosting) to allow the GAN model to leverage distributed structure of data during the training phase and generate groups at equal rate during the testing phase.