# Strong statistical parity through fair synthetic data

arXiv: https://arxiv.org/abs/2311.03000 · HTML: https://mlsanity.com/paper/2311.03000/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 15 references that are in the map.
- First posted on arXiv 6 Nov 2023.

## Builds on (most cited first, 9 of 9)

- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [FairGAN: Fairness-aware Generative Adversarial Networks](https://mlsanity.com/paper/1805.11202/) (2018)
- [A survey on datasets for fairness-aware machine learning](https://mlsanity.com/paper/2110.00530/) (2021)
- [DECAF: Generating Fair Synthetic Data Using Causally-Aware Generative Networks](https://mlsanity.com/paper/2110.12884/) (2021)
- [Fair Regression with Wasserstein Barycenters](https://mlsanity.com/paper/2006.07286/) (2020)
- [TabFairGAN: Fair Tabular Data Generation with Generative Adversarial Networks](https://mlsanity.com/paper/2109.00666/) (2021)
- [Fast, Accurate, and Simple Models for Tabular Data via Augmented Distillation](https://mlsanity.com/paper/2006.14284/) (2020)
- [Representative & Fair Synthetic Data](https://mlsanity.com/paper/2104.03007/) (2021)
- [Repairing Regressors for Fair Binary Classification at Any Decision Threshold](https://mlsanity.com/paper/2203.07490/) (2022)

## Built on by (most cited first, 6 of 6)

- [TabularARGN: A Flexible and Efficient Auto-Regressive Framework for Generating High-Fidelity Synthetic Data](https://mlsanity.com/paper/2501.12012/) (2025)
- [Debiasing Diffusion Model: Enhancing Fairness through Latent Representation Learning in Stable Diffusion Model](https://mlsanity.com/paper/2503.12536/) (2025)
- [TABFAIRGDT: A Fast Fair Tabular Data Generator using Autoregressive Decision Trees](https://mlsanity.com/paper/2509.19927/) (2025)
- [Democratizing Tabular Data Access with an Openunicodex2013Source Syntheticunicodex2013Data SDK](https://mlsanity.com/paper/2508.00718/) (2025)
- [FairFinGAN: Fairness-aware Synthetic Financial Data Generation](https://mlsanity.com/paper/2603.05327/) (2026)
- [Variable Selection in the Context of AI Fairness](https://mlsanity.com/paper/2608.11251/) (2026)

## Abstract (arXiv)

AI-generated synthetic data, in addition to protecting the privacy of original data sets, allows users and data consumers to tailor data to their needs. This paper explores the creation of synthetic data that embodies Fairness by Design, focusing on the statistical parity fairness definition. By equalizing the learned target probability distributions of the synthetic data generator across sensitive attributes, a downstream model trained on such synthetic data provides fair predictions across all thresholds, that is, strong fair predictions even when inferring from biased, original data. This fairness adjustment can be either directly integrated into the sampling process of a synthetic generator or added as a post-processing step. The flexibility allows data consumers to create fair synthetic data and fine-tune the trade-off between accuracy and fairness without any previous assumptions on the data or re-training the synthetic data generator.