Quantum Computing for Neutrino Scattering

Student: Sam Godwood and Marina Maneyro
Supervisor: Costas Andreopoulos (UoL), Gabriel Perdue (Fermilab)
Institution: University of Liverpool

The dynamics of quantum chromodynamical (QCD) bound states are of fundamental importance. The computational complexity of QCD is a major hurdle to make progress in understanding the strongly-bound systems of interest to High Energy Physics (HEP). One proposal for approaching this problem is to simulate the dynamics of QCD in a quantum computer. However, the limited number of available qubits, extremely high effective qubit connectivity requirements in field theory, and noisy gates limit the simulation in near-term systems. Further, the fact that QCD must be translated into a spin model limits the QCD models that can be studied. Different encoding schemes may be utilized to study a more generalized QCD system at the cost of increased complexity.

Qudit based systems are a promising future platform for these simulations. A qudit is an N-level generalization of the qubit (which has 2 logical levels). The recent advancements in circuit QED (cQED) allows the qudit platform to be utilized for computation with some natural advantages in terms of simulating bosonic systems and in creating quantum gates with high effective connectivity between levels. The available qudit gates that can be experimentally realized, such as phase gates or displacement gates yield a potential mechanism for lattice field theory simulation. The quantum encoding in qudits may be amplitude based, or a new encoding scheme can be devised where a particle can be encoded with the help of coherent states of light.

One interesting possibility for the simulation is the connection between entanglement entropy and the bound state and confinement in QCD. The analogues of “fermion” in qudit based systems can be realized with different arrangements of the amplitudes of the Fock states in cQED systems. Then, quantum time evolution can be applied to observe the scattering process between two fermions. Some of these problems have been studied using qubit-based quantum information processors, but we do not yet know if there may be practical advantages in using a qudit-based processor. Potential advantages may be found in the effective circuit depth, connectivity, and gate-time to coherence ratios. The entanglement entropy is straightforward to measure in cQED systems, making qudit-based platforms attractive candidates for study on the path to QCD simulation.

The Superconducting Quantum Materials and Systems Center (SQMS Center), hosted at Fermi National Accelerator Laboratory (FNAL, or Fermilab) is building a qudit-based quantum information processor that leverages world-leading expertise in superconducting devices to construct high-Q superconducting radio-frequency (SRF) cavities with the potential for revolutionary coherence times in the quantum regime. Furthermore, the SQMS Center partners with leading qubit-based computing groups, like Rigetti Computing, to develop quantum algorithms for science applications.

The PhD project will focus on 3 areas: (1) pushing the boundaries for understanding entanglement entropy on qubit-based quantum information processors using Rigetti’s quantum computing platform, (2) exploring the problem of entanglement entropy on qudit-based information processors, both theoretically in terms of gate optimization and circuit design, and from a co-design perspective emphasizing potential hardware development paths in terms of optimality for quantum simulation problems, and (3) attacking the problem of fermion scattering from a “top-down” phenomenological perspective through deep inelastic scattering (DIS) and hadronization studies designed to contextualize the quantum simulation work and connect it to intermediate-term physics goals at experiments like DUNE.

The project will leverage facilities and expertise at SQMS to drive progress on quantum simulation problems of interest to neutrino scattering. Facilities support includes access to quantum hardware (both qubit and qudit-based platforms). Expertise support includes access to both quantum programming experts and to theorists working in quantum simulation for field theory problems in HEP. Throughout the project you will have access to targeted training in data science provided by the University of Liverpool with the Centre for Doctoral Training LIV.INNO. You will also be given the opportunity to carry out an industry placement of six months to broaden your wider research and career skills.

The project is fully funded by the Fermilab SQMS center and the University of Liverpool and will be carried out over 48 months. You will spend years 1 and 4 in the UK, and be based at Fermilab during years 2 and 3. A mandatory 6-months industry placement forms part of the project.