Using Restricted Boltzmann Machines to Model Molecular Geometries

TL;DR

This paper introduces a novel RBM-based approach to model molecular geometries by training on ab initio data, enabling efficient prediction of molecular configurations, with a focus on Gaussian RBMs for small molecules.

Contribution

The paper presents a new RBM architecture based on Tanh activation and demonstrates its effectiveness in modeling small molecules like water and ethane.

Findings

Gaussian RBMs successfully model small molecules

Comparison shows Tanh RBMs outperform other activation functions

Method offers a faster alternative to traditional quantum calculations

Abstract

Precise physical descriptions of molecules can be obtained by solving the Schrodinger equation; however, these calculations are intractable and even approximations can be cumbersome. Force fields, which estimate interatomic potentials based on empirical data, are also time-consuming. This paper proposes a new methodology for modeling a set of physical parameters by taking advantage of the restricted Boltzmann machine's fast learning capacity and representational power. By training the machine on ab initio data, we can predict new data in the distribution of molecular configurations matching the ab initio distribution. In this paper we introduce a new RBM based on the Tanh activation function, and conduct a comparison of RBMs with different activation functions, including sigmoid, Gaussian, and (Leaky) ReLU. Finally we demonstrate the ability of Gaussian RBMs to model small molecules such as water and ethane.