Convex space learning for tabular synthetic data generation

2024-07-13

arXiv.org

AI Summary

This study addresses the fundamental challenge of balancing privacy preservation and data utility in clinical data sharing. We propose NextConvGeN—the first deep generative model to extend convex-space learning to full-table synthetic data generation. NextConvGeN introduces a neighborhood-driven paradigm, leveraging a GAN framework to model local neighborhood convex hulls while jointly enforcing geometric constraints and adversarial training, thereby enhancing fidelity without compromising privacy. Extensive evaluation across ten biomedical datasets demonstrates that NextConvGeN consistently outperforms five state-of-the-art methods: it achieves an average 4.2% improvement in classification F1-score and a 5.8% gain in clustering Adjusted Rand Index (ARI). All utility metrics rank among the best reported, confirming its effectiveness for downstream clinical research, machine learning model development, and clinical decision support.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Privacy
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Generating synthetic samples from the convex space of the minority class is a popular oversampling approach for imbalanced classification problems. Recently, deep-learning approaches have been successfully applied to modeling the convex space of minority samples. Beyond oversampling, learning the convex space of neighborhoods in training data has not been used to generate entire tabular datasets. In this paper, we introduce a deep learning architecture (NextConvGeN) with a generator and discriminator component that can generate synthetic samples by learning to model the convex space of tabular data. The generator takes data neighborhoods as input and creates synthetic samples within the convex space of that neighborhood. Thereafter, the discriminator tries to classify these synthetic samples against a randomly sampled batch of data from the rest of the data space. We compared our proposed model with five state-of-the-art tabular generative models across ten publicly available datasets from the biomedical domain. Our analysis reveals that synthetic samples generated by NextConvGeN can better preserve classification and clustering performance across real and synthetic data than other synthetic data generation models. Synthetic data generation by deep learning of the convex space produces high scores for popular utility measures. We further compared how diverse synthetic data generation strategies perform in the privacy-utility spectrum and produced critical arguments on the necessity of high utility models. Our research on deep learning of the convex space of tabular data opens up opportunities in clinical research, machine learning model development, decision support systems, and clinical data sharing.

Problem

• Generating synthetic tabular data via convex space learning.
• Improving classification and clustering with synthetic samples.
• Evaluating privacy and utility in synthetic data strategies.

Innovation

• Deep learning for convex space
• Generator-discriminator architecture
• Synthetic tabular data generation