Enhanced Simulation of Collision Events Using Quantum GANs for Jet Images Generation

Generative Adversarial Networks (GANs) have revolutionized unsupervised machine learning with their ability in generating identical data. Their applications range from arts to sophisticated systems such as High Energy Physics (HEP). Jets—particle showers produced by high-energy collisions provide a window into the behavior of gluons and quarks, the fundamental constituents of matter, in the domain of HEP. However, due to the huge amounts of data produced by experiments like the Large Hadron Collider (LHC), typical GANs have computational and scalability issues. Enter Quantum Generative Adversarial Networks (QGANs), which harness the power of quantum physics to promise faster training times while consuming less resources. This research digs into the complexities of QGANs and their use in jet reconstruction in HEP. This study optimizes the quality of generated images by utilizing a Quantum GAN architecture with quantum fidelity measurements. QGANs not only provides a supplement to established simulation tools like Pythia, but it also points at potential nuances and patterns that traditional simulations may ignore. Our quantum simulation results show encouraging results in jet image production and illustrate the promise of quantum techniques in interpreting the huge data landscapes of HEP.