Molecular dynamics simulations provide theoretical insight into the microscopic behavior of condensed-phase materials and, as a predictive tool, enable computational design of new compounds. However, because of the large spatial and temporal scales of thermodynamic and kinetic phenomena in materials, atomistic simulations are often computationally infeasible. Coarse-graining methods allow larger systems to be simulated by reducing their dimensionality, propagating longer timesteps, and averaging out fast motions. Coarse-graining involves two coupled learning problems: defining the mapping from an all-atom representation to a reduced representation, and parameterizing a Hamiltonian over coarse-grained coordinates. We propose a generative modeling framework based on variational auto-encoders to unify the tasks of learning discrete coarse-grained variables, decoding back to atomistic detail, and parameterizing coarse-grained force fields. The framework is tested on a number of model systems including single molecules and bulk-phase periodic simulations.

分子动力学模拟为凝相材料的微观行为提供了理论上的见识，并且作为一种预测工具，可以进行新化合物的计算设计。但是，由于材料中热力学和动力学现象的时空尺度很大，因此原子模拟通常在计算上是不可行的。粗粒度方法可通过减小大型系统的尺寸，传播更长的时间步长并平均快速运动来模拟大型系统。粗粒度涉及两个耦合的学习问题：定义从全原子表示到简化表示的映射，以及在粗粒度坐标上对哈密顿量进行参数化。我们提出了一个基于变分自动编码器的生成建模框架，以统一学习离散离散粗粒度变量，解码回原子细节和参数化粗粒度力场的任务。该框架已在许多模型系统上进行了测试，包括单分子和本体相周期性模拟。