Learning molecular energies using localized graph kernels

TL;DR

This paper introduces GRAPE, a graph kernel-based machine learning method that naturally encodes physical symmetries to accurately predict molecular energies efficiently.

Contribution

The paper presents a novel graph theory-based approach, GRAPE, that inherently incorporates physical symmetries into machine learning models for molecular energy prediction.

Findings

Achieves chemical accuracy in energy predictions

Outperforms previous models on standard datasets

Flexible framework with potential for extensions

Abstract

Recent machine learning methods make it possible to model potential energy of atomic configurations with chemical-level accuracy (as calculated from ab-initio calculations) and at speeds suitable for molecular dynam- ics simulation. Best performance is achieved when the known physical constraints are encoded in the machine learning models. For example, the atomic energy is invariant under global translations and rotations, it is also invariant to permutations of same-species atoms. Although simple to state, these symmetries are complicated to encode into machine learning algorithms. In this paper, we present a machine learning approach based on graph theory that naturally incorporates translation, rotation, and permutation symmetries. Specifically, we use a random walk graph kernel to measure the similarity of two adjacency matrices, each of which represents a local atomic environment. This Graph Approximated Energy (GRAPE) approach is flexible and admits many possible extensions. We benchmark a simple version of GRAPE by predicting atomization energies on a standard dataset of organic molecules.