Lattice simulations are an important tool to compute nonperturbative results in quantum field theories such as QCD. The most time-consuming part in lattice calculations is the solution of the Dirac equation for a given SU(3) gauge field. In the interesting physical limits, critical slowing down occurs, which can be overcome by state-of-the-art multigrid methods. We introduce gauge-equivariant neural networks that can learn the general paradigms of a multigrid. These networks can perform equally well as standard multigrids but are more general and therefore have the potential to address a larger range of research questions.
Hybrid access via ZOOM:
https://lmu-munich.zoom.us/j/98457332925?pwd=TWc3V1JkSHpyOTBPQVlMelhuNnZ1dz09
Meeting ID: 984 5733 2925
Passcode: 979953
Peter Thirolf (LMU) / Norbert Kaiser (TUM)