Towards Synthesizing High-Dimensional Tabular Data with Limited Samples

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Zuqing, Gan, Junhao, Qi, Jianzhong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918194918195200
author Li, Zuqing
Gan, Junhao
Qi, Jianzhong
author_facet Li, Zuqing
Gan, Junhao
Qi, Jianzhong
contents Diffusion-based tabular data synthesis models have yielded promising results. However, when the data dimensionality increases, existing models tend to degenerate and may perform even worse than simpler, non-diffusion-based models. This is because limited training samples in high-dimensional space often hinder generative models from capturing the distribution accurately. To mitigate the insufficient learning signals and to stabilize training under such conditions, we propose CtrTab, a condition-controlled diffusion model that injects perturbed ground-truth samples as auxiliary inputs during training. This design introduces an implicit L2 regularization on the model's sensitivity to the control signal, improving robustness and stability in high-dimensional, low-data scenarios. Experimental results across multiple datasets show that CtrTab outperforms state-of-the-art models, with a performance gap in accuracy over 90% on average.
format Preprint
id arxiv_https___arxiv_org_abs_2503_06444
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Synthesizing High-Dimensional Tabular Data with Limited Samples
Li, Zuqing
Gan, Junhao
Qi, Jianzhong
Machine Learning
Artificial Intelligence
Databases
Diffusion-based tabular data synthesis models have yielded promising results. However, when the data dimensionality increases, existing models tend to degenerate and may perform even worse than simpler, non-diffusion-based models. This is because limited training samples in high-dimensional space often hinder generative models from capturing the distribution accurately. To mitigate the insufficient learning signals and to stabilize training under such conditions, we propose CtrTab, a condition-controlled diffusion model that injects perturbed ground-truth samples as auxiliary inputs during training. This design introduces an implicit L2 regularization on the model's sensitivity to the control signal, improving robustness and stability in high-dimensional, low-data scenarios. Experimental results across multiple datasets show that CtrTab outperforms state-of-the-art models, with a performance gap in accuracy over 90% on average.
title Towards Synthesizing High-Dimensional Tabular Data with Limited Samples
topic Machine Learning
Artificial Intelligence
Databases
url https://arxiv.org/abs/2503.06444