DSF-GAN: DownStream Feedback Generative Adversarial Network
This work addresses the problem of low utility in synthetic data generation for tabular datasets, which is incremental as it builds on existing GAN architectures.
The paper tackled the challenge of generating synthetic tabular data with high utility by proposing DSF-GAN, which incorporates feedback from a downstream prediction model during training, and demonstrated improved model performance on synthetic samples compared to a baseline GAN without feedback.
Utility and privacy are two crucial measurements of the quality of synthetic tabular data. While significant advancements have been made in privacy measures, generating synthetic samples with high utility remains challenging. To enhance the utility of synthetic samples, we propose a novel architecture called the DownStream Feedback Generative Adversarial Network (DSF-GAN). This approach incorporates feedback from a downstream prediction model during training to augment the generator's loss function with valuable information. Thus, DSF-GAN utilizes a downstream prediction task to enhance the utility of synthetic samples. To evaluate our method, we tested it using two popular datasets. Our experiments demonstrate improved model performance when training on synthetic samples generated by DSF-GAN, compared to those generated by the same GAN architecture without feedback. The evaluation was conducted on the same validation set comprising real samples. All code and datasets used in this research will be made openly available for ease of reproduction.