🤖 AI Summary
This work addresses the challenge of generating high-quality synthetic tabular data for severely imbalanced datasets, where existing methods often fail to preserve both fidelity and utility for minority classes in downstream tasks. The authors propose the CTTVAE+TBS framework, which integrates a conditional Transformer-based variational autoencoder with a class-aware triplet boundary loss to restructure the latent space—enhancing intra-class compactness and inter-class separability. Additionally, an adaptive training sampling mechanism dynamically increases minority class exposure during training. Extensive experiments on six real-world datasets demonstrate that the proposed method significantly outperforms baseline approaches, achieving high data fidelity while substantially improving downstream task performance for minority classes—even surpassing models trained on the original imbalanced data—and effectively bridging the privacy gap between interpolation-based and deep generative methods.
📝 Abstract
Generating synthetic tabular data under severe class imbalance is essential for domains where rare but high-impact events drive decision-making. However, most generative models either overlook minority groups or fail to produce samples that are useful for downstream learning. We introduce CTTVAE, a Conditional Transformer-based Tabular Variational Autoencoder equipped with two complementary mechanisms: (i) a class-aware triplet margin loss that restructures the latent space for sharper intra-class compactness and inter-class separation, and (ii) a training-by-sampling strategy that adaptively increases exposure to underrepresented groups. Together, these components form CTTVAE+TBS, a framework that consistently yields more representative and utility-aligned samples without destabilizing training. Across six real-world benchmarks, CTTVAE+TBS achieves the strongest downstream utility on minority classes, often surpassing models trained on the original imbalanced data while maintaining competitive fidelity and bridging the gap for privacy for interpolation-based sampling methods and deep generative methods. Ablation studies further confirm that both latent structuring and targeted sampling contribute to these gains. By explicitly prioritizing downstream performance in rare categories, CTTVAE+TBS provides a robust and interpretable solution for conditional tabular data generation, with direct applicability to industries such as healthcare, fraud detection, and predictive maintenance where even small gains in minority cases can be critical.