Lattice QCD towards the Exascale Computing Era

Andrea Shindler, Michigan State University
Wednesday, Apr 08, 4:10 PM - Nuclear Science Seminar
Online via Zoom

Abstract:  The Standard Model of Particle Physics (SM) has been widely successful in describing the measured particle spectrum and composition of matter ranging from quarks and gluons to multi-hadron systems. Such systems constitute approximately 5% of the observable matter-energy within the Universe. Yet the theory alone cannot explain the origins of the remaining 95% of matter and energy, dubbed "Dark Matter" and "Dark Energy", respectively. Further, the Standard Model does not provide the requisite amount of CP violation to account for the observed matter-antimatter asymmetry. Therefore any physical description of such phenomena requires a theory that goes beyond the Standard Model (BSM), while at the same time encompassing the Standard Model and its predictions related to ordinary matter. The electric dipole moments (EDM) of the nucleon is a very sensitive probe of CP violation and the current bound on the neutron EDM strongly constrains many BSM physics scenarios. The considerable challenges presented by the calculation of the nucleon EDM from Lattice QCD can only be addressed combining Exascale computing capacities with the development of new algorithms, and new theoretical and numerical techniques. I will present our Lattice QCD results of the first determination of the theta-term contribution to the nucleon EDM and towards a complete determination of all the CP-violating contributions to the nucleon EDM. I will emphasize the role played by a novel theoretical tool, the Gradient Flow, and by a novel noise reduction technique we recently introduced. I will also present our new algorithmic developments based on Machine Learning techniques and the perspective for lattice QCD calculations using Quantum Computers.