Overview

An exploration of Quantum Generative Adversarial Networks for generating and analyzing image structures that are difficult to model with classical GANs alone.

Focus

  • Designing generator and discriminator circuits suitable for limited qubit budgets
  • Comparing sample quality against a reduced classical GAN baseline
  • Studying how circuit depth and encoding choice affect generated structure

Outcome

The experiments and circuit definitions are documented in the repository, with plots used to trace the behaviour of the generated samples.