Abstract

This talk introduces key ideas in quantum machine learning, with a focus on hybrid quantum–classical generative models. I will present our recent work on QTabGAN, a hybrid quantum–classical GAN designed for realistic tabular data synthesis. The model uses variational quantum circuits to learn expressive probability distributions, combined with classical neural networks to generate structured tabular data. Alongside the technical details of the model and experimental results, the talk will also discuss the broader role of quantum computing in generative modelling, its potential advantages for complex data distributions, and practical challenges for near-term quantum hardware. https://arxiv.org/pdf/2602.12704