Quantum optimization and machine learning
Quantum computers can manipulate large amounts of data, such as whole probability distributions, at once. This makes them incredibly powerful, enabling Fermilab scientists to use them for certain complex physics problems like constrained optimization problems. They can also help with machine learning problems, making connections between inputs and learning how strong those connections should be. For instance, Fermilab scientists are using artificial intelligence to enhance the performance of networks of quantum sensors to search for new particles and new physics.
Scientists frequently use optimization algorithms to analyze high-energy physics data. These algorithms perform tasks such as fitting waveforms from particle detectors to templates to determine particle energies and finding set of potential paths that are closest to a particle’s true trajectory. Fermilab is developing quantum algorithms to solve very basic problems and then scaling them up to apply them to the kinds of problems that need to be solved in high-energy physics.
Physics theory applications
Fermilab has a strong research program in quantum theory and applications. Quantum theory is the key to understanding everything in the realm of quantum science and technology. Theorists are studying ways to move quantum science forward and use quantum devices to solve problems too difficult for even the most powerful supercomputers. Fermilab theorists aim to explore and further develop connections between quantum science, which involves subatomic particles and forces, and quantum field theories, which provide the foundation for particle physics studies.
Quantum field theories allow theorists to describe and compute phenomena in a vast range of scale, including examples from the theory of quantum electrodynamics, the effective field theories of the strong and weak interactions at low energies, the gauge theories of the Standard Model of particle physics, and possible extensions beyond the Standard Model. Programs that simulate quantum theories and other complex phenomena are extremely resource intensive because of the many computations required to keep track of complex data like entanglement and superposition. Quantum computers are perfectly suited to this work.
Fermilab scientists are also investigating approaches to reducing errors in simulation of quantum field theories using specialized techniques. They are especially interested in simulating quantum chromodynamics and aim to leverage Fermilab’s expertise in lattice QCD computations to inform and improve quantum simulation.
Large-scale simulations of quantum systems on high-performance computing
Large-scale simulation of quantum computers is akin to simulation of high-energy physics interactions: Both must manage a vast array of variables; both organize their inputs and outputs similarly; and for both, the simulations must be analyzed and consolidated into results. Fermilab scientists, in collaboration with scientists at Argonne National Laboratory, are using high-energy physics tools to produce and analyze simulations using high-performance computers at the Argonne Leadership Computing Facility.
They are simulating the operation of a qubit device that uses superconducting cavities from particle accelerators to maintain quantum information over a relatively long time. Their results will determine the device’s impact on high-energy physics algorithms using a quantum simulator developed at Argonne National Laboratory.
Quantum computation of fermion and boson systems
Fermilab scientists are developing ways to use quantum computers to simulate some of the most basic particle interactions, including those that hold our universe together: fermions, the building blocks of matter, and bosons, the field particles that pull on particles of matter. Scientists have already made strides in modeling systems composed of fermions using quantum algorithms. It’s been more challenging to achieve this for bosons. Fermilab has found a way to do it on quantum computers and are further developing the model. They represent the bosons as a system of harmonic oscillators. Harmonic oscillation occurs everywhere in nature: the motion of a spring bobbing up and down, the vibration of a plucked string. In quantum mechanics, harmonic oscillator motion is described by wave functions. It is so well understood, they can be precisely modeled, opening the door to realistically simulating the subatomic realm.