Bayesian Generative Adversarial Networks via Gaussian Approximation for Tabular Data Synthesis

TL;DR

This paper introduces GACTGAN, a Bayesian GAN variant using Gaussian approximation to synthesize tabular data more effectively and efficiently than existing methods, with improved data quality and privacy features.

Contribution

GACTGAN integrates SWAG for Bayesian approximation in CTGAN, reducing computational costs while enhancing data synthesis quality for tabular data.

Findings

GACTGAN outperforms CTGAN in data quality and privacy preservation.

GACTGAN reduces computational overhead compared to MCMC-based Bayesian GANs.

GACTGAN better maintains tabular structure and statistical properties.

Abstract

Generative Adversarial Networks (GAN) have been used in many studies to synthesise mixed tabular data. Conditional tabular GAN (CTGAN) have been the most popular variant but struggle to effectively navigate the risk-utility trade-off. Bayesian GAN have received less attention for tabular data, but have been explored with unstructured data such as images and text. The most used technique employed in Bayesian GAN is Markov Chain Monte Carlo (MCMC), but it is computationally intensive, particularly in terms of weight storage. In this paper, we introduce Gaussian Approximation of CTGAN (GACTGAN), an integration of the Bayesian posterior approximation technique using Stochastic Weight Averaging-Gaussian (SWAG) within the CTGAN generator to synthesise tabular data, reducing computational overhead after the training phase. We demonstrate that GACTGAN yields better synthetic data compared to CTGAN, achieving better preservation of tabular structure and inferential statistics with less privacy risk. These results highlight GACTGAN as a simpler, effective implementation of Bayesian tabular synthesis.