用贝叶斯方法改进生成对抗网络，提升表格数据合成质量与隐私保护。
Bayesian Generative Adversarial Networks via Gaussian Approximation for Tabular Data Synthesis
- 在CTGAN生成器中引入SWAG进行后验近似，降低计算开销。
- 合成数据更保真结构与统计特征，隐私风险更低。
- 适合关注数据隐私与生成质量的机器学习研究者。
生成对抗网络（GAN）被广泛用于混合表格数据的合成。条件表格式GAN（CTGAN）是最流行的变体，但在风险-效用权衡上表现不佳。贝叶斯GAN在表格数据中研究较少，多用于图像和文本等非结构化数据。现有贝叶斯GAN主要依赖马尔可夫链蒙特卡洛（MCMC），计算成本高，尤其在权重存储方面。本文提出高斯近似CTGAN（GACTGAN），将随机权重平均-高斯（SWAG）技术融入CTGAN生成器，实现训练后低计算开销的贝叶斯后验近似。实验表明，相比CTGAN，GACTGAN生成的数据在保持表格结构与推断统计特性方面更优，且隐私风险更低。结果证明GACTGAN是贝叶斯表格数据合成的一种更简单有效的实现。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。