A Hessian-Based Assessment of Atomic Forces for Training Machine Learning Interatomic Potentials

Abstract

In recent years, many types of machine learning potentials (MLPs) have been introduced, which are able to represent high-dimensional potential-energy surfaces (PES) with close to first-principles accuracy. Most current MLPs rely on atomic energy contributions given as a function of the local chemical environments. Frequently, in addition to total energies also atomic forces are used to construct the potentials, as they provide detailed local information about the PES. Since many systems are too large for electronic structure calculations, obtaining reliable reference forces from smaller subsystems like molecular fragments or clusters can substantially simplify the construction of the training sets. Here, we propose a method to determine structurally converged molecular fragments providing reliable atomic forces based on an analysis of the Hessian. The method, which serves as a locality test and allows to estimate the importance of long-range interactions, is illustrated for a series of molecular model systems and the metal-organic framework MOF-5 as an example for a complex organic-inorganic hybrid material.

15 pages, 13 figures excluding the supporting information, submitted to JCP

References in corpus (5)

• ANI-1: An extensible neural network potential with DFT accuracy at force field computational cost
• A Spectral Analysis Method for Automated Generation of Quantum-Accurate Interatomic Potentials
• Machine learning for molecular simulation
• Machine-learning based interatomic potential for amorphous carbon
• Comparing the Accuracy of High-Dimensional Neural Network Potentials and the Systematic Molecular Fragmentation Method: A Benchmark Study for all-trans Alkanes