Machine Learning in Equation of State and Transport Modeling at Extreme Conditions

Explore opportunities for machine learning in equation of state and transport modeling at extreme conditions in this 49-minute conference talk. Delve into the challenges of understanding macroscopic material evolution in Jovian planet modeling and inertial confinement fusion design. Discover how machine learning can enhance traditional density functional theory (DFT) and quantum Monte Carlo (QMC) workflows, potentially improving accuracy and reducing costs. Examine topics such as the National Ignition Facility, compensating errors, uncertainty quantification, multifidelity modeling, and spectral neighbor analysis potential. Gain insights into improving wave functions, Jastrow factors, and optimal pair potentials for more accurate equations of state and transport properties in extreme conditions.

Intro
National Ignition Facility
Compensating Errors
Equation of State
QMC
Uncertainty Quantification
Aluminum
Real Space QMC
Aluminum Project
Multifidelity Modeling
Onsatsa
Improving wave functions
Improving Jastro
Spectral Neighbor Analysis Potential
Performance
Initial implementation
Potentials
Optimal Pair Potential
Dynamical Properties
Conclusion