Quantum Computing Methods for High Energy Physics

10 April - 5 May 2023

Christian Bauer, Ignacio Cirac, Marcello Dalmonte, Zohreh Davoudi, Henry S. Lamm, Michael Spannowsky

This four-week programme brings together world-leading experts working at the intersection of quantum-information sciences (QIS) and high-energy physics (HEP), with a focus on quantum simulation, quantum machine learning, and tensor networks. Each represents an area with outstanding problems, or where imminent significant progress is anticipated. Quantum algorithms are predicted to outperform classical algorithms, and quantum hardware continues to improve in scale, reliability, and applicability. With advances in theory, algorithm, and hardware over the past decade, the interest in applying QIS paradigms to answer questions in HEP has surged. Quantum simulation of HEP will enable studies of large entangled Hilbert spaces and offers a solution to the sign problem, situations where classical methods appear insufficient. The physics applications span many HEP topics: realtime dynamics of matter in and out of equilibrium in collider experiments and early universe, nonperturbative inputs into event generators for the LHC and beyond, predicting the QCD equation of state for LIGO and astrophysics, and insights into quantum gravity and black-hole physics. In recent years, progress has been made in finding efficient formulations, realistic analog proposals, nearand far-term digital algorithms, and small hardware demonstrations. Developing a clear understanding of where the boundary of quantum advantage lies in HEP simulations is an objective of the community in the coming years.Today, machine learning is a vital tool for big data analysis. Consequently, quantum machine learning has the potential to further enhance, speed up or altogether change the process of data analysis. Existing applications of quantum machine learning to high-energy physics include supervised classification tasks for reconstructed objects or processes, e.g. signal discrimination, anomaly detection methods, and particle track reconstruction. Tensor networks can be thought of as a data compression protocol to describe quantum systems by representing wave functions through a network of properly dovetailed interconnected building blocks. These networks are found to provide accurate encodings of the relevant properties, including quantum entanglement: they have been shown to provide insights in regimes where Monte Carlo simulations are not always applicable, such as finite-density of fermions and real time dynamics, while facing challenges in higher dimensional systems. These and related developments may allow researchers to apply tensor networks to a wide class of problems in high-energy physics, and to take advantage of them in benchmarking and guiding quantum-simulation protocols.