Speaker
Description
This talk presents two studies conducted at ParityQC on HEP applications, covering both theoretical and experimental perspectives. The first study (arXiv:2606.28236) focuses on simulating electromagnetic particle showers in calorimeters, a computationally intensive task due to the complexity of energy deposition distributions. Quantum computing offers a potential alternative, but existing approaches often require binarizing data or encoding it into gate angles, which can distort non-binary distributions. While Instantaneous Quantum Polynomial (IQP) circuits have shown promise for quantum generative modeling, they have mainly been applied to binary data. In our work, we extend parameterized IQP circuits to qudits, designing a qudit-based quantum circuit and correspondingly adapt the loss function. The method, validated on particles datasets, can be extended to other applications that utilize quantum generative machine learning for non-binary data. The first-part of the talk is supplemented by the poster ‘Qudit extension of parameterized IQP circuits: A generative quantum machine learning approach to calorimeter data’.
The second study (arXiv:2605.12600) introduces two techniques for efficient fermionic quantum simulation: a dynamically reconfigurable Jordan–Wigner transformation that preserves locality in higher-dimensional lattices, and an optimized fermionic routing scheme for implementing non-local interactions. This study enables efficient simulation, with asymptotically optimal resource scaling, of fermionic systems with applications to the Fermi–Hubbard model.