Adversarial-residual-coarse-graining: Applying machine learning theory to systematic molecular coarse-graining
Adversarial-residual-coarse-graining: Applying machine learning theory to systematic molecular coarse-graining
复制标题
DOI：
10.1063/1.5097559
复制
发表时间：
2019-09-28
影响因子：
4.4
通讯作者：
Voth, Gregory A.
中科院分区：
文献类型：
--
作者：
Durumeric, Aleksander E. P.;Voth, Gregory A.
我们利用机器学习中分子粗粒度（CG）方法和隐式生成模型之间的联系来描述系统分子粗粒度（CG）的新框架。重点放在包含生成对抗网络的形式主义上。该方法支持多种模型参数化策略，其中一些策略与以前的CG方法相似。我们证明了所得到的框架可以严格地参数化CG模型，其中包含与参考原子系统（称为虚拟站点）没有规定连接的CG站点；然而，这一优势被缺乏在虚拟CG站点上集成后获得的分辨率上的CG哈密顿量的封闭形式表达式所抵消。在这些方法理想地返回与相对熵最小化CG相同的参数，但传统的相对熵最小化CG优化方程不适用的情况下，提供了计算实例。
We utilize connections between molecular coarse-graining (CG) approaches and implicit generative models in machine learning to describe a new framework for systematic molecular CG. Focus is placed on the formalism encompassing generative adversarial networks. The resulting method enables a variety of model parameterization strategies, some of which show similarity to previous CG methods. We demonstrate that the resulting framework can rigorously parameterize CG models containing CG sites with no prescribed connection to the reference atomistic system (termed virtual sites); however, this advantage is offset by the lack of a closed-form expression for the CG Hamiltonian at the resolution obtained after integration over the virtual CG sites. Computational examples are provided for cases in which these methods ideally return identical parameters as relative entropy minimization CG but where traditional relative entropy minimization CG optimization equations are not applicable.