Dynamic electronic processes play an important role in the functionality of many electronic devices. Multiscale simulation approaches link quantum and device scales aiming at an understanding of the emergence of macroscopically observable physical properties such as conductivity or carrier mobility. In many cases though, the computational cost of such models renders the macroscale inaccessible. I will discuss these problems and how embedding Machine Learning into the simulation framework helps to overcome its limitations, paving the way to extracting physical laws of macroscopic behavior from the fundamental equations governing the microscale.
Back to Workshop III: Validation and Guarantees in Learning Physical Models: from Patterns to Governing Equations to Laws of Nature