Understanding the macroscopic evolution of materials at extreme conditions is a critical component of Jovian planet modeling and inertial confinement fusion design. This is unfortunately a massively multiscale problem where one traditionally uses density functional theory (DFT) and quantum Monte Carlo (QMC) to construct equation of state and transport tables, which then serve as inputs to hydrodynamic simulations. In this talk, I will discuss several opportunities for machine learning to augment our traditional DFT/QMC workflow, potentially allowing us to achieve more accurate equations of state and transport properties, with more physical insights, at reduced cost to current standard practices.
Back to Workshop IV: Monte Carlo and Machine Learning Approaches in Quantum Mechanics
Copyright. All Rights Reserved.