Comparing molecules and solids across structural and alchemical space

TL;DR

This paper introduces a unified framework combining local atomic descriptors and a regularized matching approach to compare molecules and solids, enabling effective navigation of structural and chemical space and accurate property prediction.

Contribution

It presents a novel method that integrates SOAP descriptors with REMatch to compare entire molecular and periodic structures, advancing machine learning applications in materials science.

Findings

Achieved mean absolute error below 1 kcal/mol in atomization energy predictions.

Developed powerful metrics for structural and chemical similarity across diverse compounds.

Unified approach facilitates exploration of complex materials and molecular spaces.

Abstract

Evaluating the (dis)similarity of crystalline, disordered and molecular compounds is a critical step in the development of algorithms to navigate automatically the configuration space of complex materials. For instance, a structural similarity metric is crucial for classifying structures, searching chemical space for better compounds and materials, and driving the next generation of machine-learning techniques for predicting the stability and properties of molecules and materials. In the last few years several strategies have been designed to compare atomic coordination environments. In particular, the Smooth Overlap of Atomic Positions (SOAP) has emerged as an elegant framework to obtain translation, rotation and permutation-invariant descriptors of groups of atoms, driven by the design of various classes of machine-learned inter-atomic potentials. Here we discuss how one can combine such local descriptors using a Regularized Entropy Match (REMatch) approach to describe the similarity of both whole molecular and bulk periodic structures, introducing powerful metrics that enable the navigation of alchemical and structural complexity within a unified framework. Furthermore, using this kernel and a ridge regression method we can predict atomization energies for a database of small organic molecules with a mean absolute error below 1kcal/mol, reaching an important milestone in the application of machine-learning techniques to the evaluation of molecular properties.