Neural Network Potentials: A Concise Overview of Methods

2021/07/08 by Emir Kocer, Tsz Wai Ko, Kocer, Emir +3 · 19 citations
Chemistry · Computer Science · Engineering · Materials Science · Mathematics · Physics and Astronomy · #Artificial intelligence #Artificial neural network #Computational Drug Discovery Methods #Computer science #Curse of dimensionality #Electrochemical Analysis and Applications #Engineering #Feature (linguistics) #Field (mathematics) #Machine Learning in Materials Science #Machine learning #Mathematics #Range (aeronautics) #physics.chem-ph

paper · pdf · doi:10.1146/annurev-physchem-082720-034254

published in Annual Review of Physical Chemistry 73(1), 163-186 (Annual Reviews)

arxiv created 2021/07/08 · arxiv updated 2021/07/09 · openalex publication_date 2022/01/04 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/06

Abstract

In the past two decades, machine learning potentials (MLPs) have reached a level of maturity that now enables applications to large-scale atomistic simulations of a wide range of systems in chemistry, physics, and materials science. Different machine learning algorithms have been used with great success in the construction of these MLPs. In this review, we discuss an important group of MLPs relying on artificial neural networks to establish a mapping from the atomic structure to the potential energy. In spite of this common feature, there are important conceptual differences among MLPs, which concern the dimensionality of the systems, the inclusion of long-range electrostatic interactions, global phenomena like nonlocal charge transfer, and the type of descriptor used to represent the atomic structure, which can be either predefined or learnable. A concise overview is given along with a discussion of the open challenges in the field.

Citations

• Gaussian approximation potentials: A brief tutorial introduction
• Die Berechnung optischer und elektrostatischer Gitterpotentiale
• Constructing high‐dimensional neural network potentials: A tutorial review
• Quantum-chemical insights from deep tensor neural networks
• SchNet – A deep learning architecture for molecules and materials
• Less is more: Sampling chemical space with active learning
• DScribe: Library of descriptors for machine learning in materials science
• Machine Learning for Molecular Simulation

Cited by

• Recent advances in stimuli-responsive framework materials: Understanding their response and searching for materials with targeted behavior
• Pushing the Accuracy Limit of Foundation Neural Network Models with Quantum Monte Carlo Forces and Path Integrals
• The Convergence Frontier: Integrating Machine Learning and High Performance Quantum Computing for Next-Generation Drug Discovery
• Long-range electrostatics for machine learning interatomic potentials is easier than we thought
• Mapping Still Matters: Coarse-Graining with Machine Learning Potentials
• Machine Learning Photodynamics Unveils a Controlled H₂ Loss Channel in Methaniminium Cation
• Field theoretic atomistics: Learning thermodynamic and variational surrogate to density functional theory
• A Fast, Accurate, and Reactive Equivariant Foundation Potential
• From QM/MM to ML/MM: A new era in multiscale modeling
• Atomistic Simulations of Oxide-Water Interfaces using Machine Learning Potentials
• Unconventional superconductivity from lattice quantum disorder
• Machine-Learning Potentials for Efficient Simulations of Anisotropic Colloids
• Insights into the Structure and Dynamics of Water at Co₃O₄(001) Using a High-Dimensional Neural Network Potential
• Autonomous thermodynamically informed database generation for machine-learned interatomic potentials and application to magnesium
• Hybrid Quantum--Classical Machine Learning Potential with Variational Quantum Circuits
• On-the-Fly Fine-Tuning of Foundational Neural Network Potentials: A Bayesian Neural Network Approach
• Machine-Learned Potentials for Solvation Modeling
• Nonadiabatic reactive scattering of hydrogen on different surface facets of copper
• HORM: A Large Scale Molecular Hessian Database for Optimizing Reactive Machine Learning Interatomic Potentials

Related

• Four Generations of High-Dimensional Neural Network Potentials
• Constructing high‐dimensional neural network potentials: A tutorial review
• Interatomic Potential in a Simple Dense Neural Network Representation
• Introduction to machine learning potentials for atomistic simulations
• The evolution of machine learning potentials for molecules, reactions and materials