Accurate force fields are necessary for predictive molecular simulations. However, developing force fields that accurately reproduce experimental properties is challenging. Here, we present a machine learning directed, multiobjective optimization workflow for force field parameterization that evaluates millions of prospective force field parameter sets while requiring only a small fraction of them to be tested with molecular simulations. We demonstrate the generality of the approach and identify multiple low-error parameter sets for two distinct test cases: simulations of hydrofluorocarbon (HFC) vapor-liquid equilibrium (VLE) and an ammonium perchlorate (AP) crystal phase. We discuss the challenges and implications of our force field optimization workflow.
@article{arxiv.2103.03208,
title = {Machine Learning Directed Optimization of Classical Molecular Modeling Force Fields},
author = {Bridgette J. Befort and Ryan S. DeFever and Garrett M. Tow and Alexander W. Dowling and Edward J. Maginn},
journal= {arXiv preprint arXiv:2103.03208},
year = {2021}
}