![[2602.12704] QTabGAN: A Hybrid Quantum-Classical GAN for Tabular Data Synthesis](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.12704] QTabGAN: A Hybrid Quantum-Classical GAN for Tabular Data Synthesis
Summary
QTabGAN introduces a hybrid quantum-classical generative adversarial network designed for synthesizing tabular data, addressing challenges like heterogeneous features and data scarcity.
Why It Matters
As data privacy concerns grow, QTabGAN offers a novel solution for generating realistic tabular data without compromising sensitive information. This advancement could enhance machine learning applications where data availability is limited, making it highly relevant in today's data-driven landscape.
Key Takeaways
- QTabGAN combines quantum computing and classical neural networks for data synthesis.
- It significantly improves performance on classification tasks, achieving up to 54.07% better results.
- The model is particularly useful in scenarios with limited or sensitive data.
Computer Science > Machine Learning arXiv:2602.12704 (cs) [Submitted on 13 Feb 2026] Title:QTabGAN: A Hybrid Quantum-Classical GAN for Tabular Data Synthesis Authors:Subhangi Kumari, Rakesh Achutha, Vignesh Sivaraman View a PDF of the paper titled QTabGAN: A Hybrid Quantum-Classical GAN for Tabular Data Synthesis, by Subhangi Kumari and 2 other authors View PDF HTML (experimental) Abstract:Synthesizing realistic tabular data is challenging due to heterogeneous feature types and high dimensionality. We introduce QTabGAN, a hybrid quantum-classical generative adversarial framework for tabular data synthesis. QTabGAN is especially designed for settings where real data are scarce or restricted by privacy constraints. The model exploits the expressive power of quantum circuits to learn complex data distributions, which are then mapped to tabular features using classical neural networks. We evaluate QTabGAN on multiple classification and regression datasets and benchmark it against leading state-of-the-art generative models. Experiments show that QTabGAN achieves up to 54.07% improvement across various classification datasets and evaluation metrics, thus establishing a scalable quantum approach to tabular data synthesis and highlighting its potential for quantum-assisted generative modelling. Comments: Subjects: Machine Learning (cs.LG); Quantum Physics (quant-ph) Cite as: arXiv:2602.12704 [cs.LG] (or arXiv:2602.12704v1 [cs.LG] for this version) https://doi.org/10.48550/arXi...
