Adversarial reverse mapping of equilibrated condensed-phase molecular structures

Mach. Learn.: Sci. Technol. 1 (2020)
Author

Stieffenhofer, Wand, Bereau

Published

2020-09-09

Doi



A tight and consistent link between resolutions is crucial to further expand the impact of multiscale modeling for complex materials. We herein tackle the generation of condensed molecular structures as a refinement—backmapping—of a coarse-grained (CG) structure. Traditional schemes start from a rough coarse-to-fine mapping and perform further energy minimization and molecular dynamics simulations to equilibrate the system. In this study we introduce DeepBackmap: A deep neural network based approach to directly predict equilibrated molecular structures for condensed-phase systems. We use generative adversarial networks to learn the Boltzmann distribution from training data and realize reverse mapping by using the CG structure as a conditional input. We apply our method to a challenging condensed-phase polymeric system. We observe that the model trained in a melt has remarkable transferability to the crystalline phase. The combination of data-driven and physics-based aspects of our architecture help reach temperature transferability with only limited training data.

from Orcid & CrossRef

This study introduces DeepBackmap: A deep neural network based approach to directly predict equilibrated molecular structures for condensed-phase systems and observes that the model trained in a melt has remarkable transferability to the crystalline phase.

from Semantic Scholar
@article{Stieffenhofer_2020,
    doi = {10.1088/2632-2153/abb6d4},
    url = {https://doi.org/10.1088%2F2632-2153%2Fabb6d4},
    year = 2020,
    month = {oct},
    publisher = {{IOP} Publishing},
    volume = {1},
    number = {4},
    pages = {045014},
    author = {Marc Stieffenhofer and Michael Wand and Tristan Bereau},
    title = {Adversarial reverse mapping of equilibrated condensed-phase molecular structures},
    journal = {Machine Learning: Science and Technology}
}
from doi2bib
Webpage PDF

from Unpaywall