Financial datasets often suffer from bias that can lead to unfair decision-making in automated systems. In this work, we propose FairFinGAN, a WGAN-based framework designed to generate synthetic financial data while mitigating bias with respect to the protected attribute. Our approach incorporates fairness constraints directly into the training process through a classifier, ensuring that the synthetic data is both fair and preserves utility for downstream predictive tasks. We evaluate our proposed model on five real-world financial datasets and compare it with existing GAN-based data generation methods. Experimental results show that our approach achieves superior fairness metrics without significant loss in data utility, demonstrating its potential as a tool for bias-aware data generation in financial applications.
@article{arxiv.2603.05327,
title = {FairFinGAN: Fairness-aware Synthetic Financial Data Generation},
author = {Tai Le Quy and Dung Nguyen Tuan and Trung Nguyen Thanh and Duy Tran Cong and Huyen Giang Thi Thu and Frank Hopfgartner},
journal= {arXiv preprint arXiv:2603.05327},
year = {2026}
}
Comments
Accepted to Special Session: Data Science: Foundations and Applications (DSFA), PAKDD 2026