Machine Learning Force Fields

By Oliver T. Unke, Stefan Chmiela, Huziel E. Sauceda, Michael Gastegger, Igor Poltavsky, Kristof T. Schütt, Alexandre Tkatchenko, K. Müller. Chemical Reviews, 2020

1,403 citations recorded by the source corpus.

In recent years, the use of machine learning (ML) in computational chemistry has enabled numerous advances previously out of reach due to the computational complexity of traditional electronic-structure methods. One of the most promising applications is the construction of ML-based force fields (FFs), with the aim to narrow the gap between the accuracy of ab initio methods and the efficiency of classical FFs. The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions. Such universal ML approximations are in principle only limited by the quality and quantity of the reference data used to train them. This review gives an overview of applications of ML-FFs and the chemical insights that can be obtained from them. The core concepts underlying ML-FFs are described in detail, and a step-by-step guide for constructing and testing them from scratch is given. The text concludes with a discussion of the challenges that remain to be overcome by the next generation of ML-FFs.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/a1f57b760b7d4c44490f0bb86ee5462c3e7c7272

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

In Materials Chemistry

In Molecular Biology

Citations in the corpus

Higher-Order Explanations of Graph Neural Networks via Relevant Walks (2020, background): “In the field of machine learning for quantum chemistry [68], [69], [70], [71], [72], GNNs have been exhibiting state of the art performance for predicting molecular properties [3], [37],”
SE(3)-equivariant prediction of molecular wavefunctions and electronic densities (2021, methodology): “The CCSD(T) method models the effects of electronic many-body interactions, which are neglected in HF theory, and is often considered to be the “gold standard” of quantum chemistry [8].”
Deep Potentials for Materials Science (2022, background): “Apart from NNP and SchNet, several other types of ML potentials were introduced; Gaussian approximation potentials (GAP) [30– 33], moment tensor potentials (MTP) [34–36], spectral neighbour analysis potentials (SNAP) [37–39], gradient-domain machine learning (GDML) [40, 41], et al [42].”