An On-the-Fly Approach to Construct Generalized Energy-Based Fragmentation Machine Learning Force Fields of Complex Systems

Abstract

An on-the-fly fragment-based machine learning (ML) approach was developed to construct the machine learning force field for large complex systems. In this approach, the energy, forces, and molecular properties of the target system are obtained by combining machine learning force fields of various subsystems with the generalized energy-based fragmentation (GEBF) approach. Using nonparametric Gaussian process (GP) model, all the force fields of subsystems are automatically generated online without data selection and parameter optimization. With the GEBF-ML force field constructed for a normal alkane, C60H122, long-time molecular dynamics (MD) simulations are performed on different sizes of alkanes, and the predicted energy, forces, and molecular properties (dipole moment) are favorably comparable with full quantum mechanics (QM) calculations. The predicted IR spectra also show excellent agreement with the direct ab initio MD results. Our results demonstrate that the GEBF-ML method provides an automatic and efficient way to build force fields for a broad range of complex systems such as biomolecules and supramolecular systems.

Computational and data driven molecular material design assisted by low scaling quantum mechanics calculations and machine learning

“…For example, an online transfer learning algorithm was developed for predicting the force elds, conformational structures, and IR spectra of polymer oligomers by training GEBF subsystems only. 44,218 In the future, the online machine learning force eld should also be extended to periodic materials and interfaces by taking innite long-range electrostatic interactions into accounts. Then, the lattice energy, crystal structures, and spectroscopic properties of periodic materials could be efficiently obtained to assist the design of novel functional materials.…”

“…3). 44,171 Such a fragmentation scheme also works well for the p-conjugated chains like polyacetylene and polyuorenols but the p-bonds cannot be cut during the fragmentation. 172,173 An auto fragmentation program was implemented for more complicated molecular aggregate or biological systems with a pre-setting truncate distance for building the subsystems in GEBF calculations.…”

“…An on-the-y GEBF ML approach was developed to construct a machine learning force eld for oligomers by employing a Gaussian process without the need for data selection and parameter optimization. 44,218 In this approach, only those small GEBF subsystems are employed to construct ML force elds automatically and efficiently. Then, the total energy, gradients, and molecular properties could be obtained from those of GEBF subsystems.…”

“…The efficiency of building the force elds from the low scaling QM-based on-line machine learning force eld is also discussed. 44,218 Finally, the perspective discusses some directions for further improvement and proposes future lines of this exciting and quickly developing eld.…”

Transferable Multi-level Attention Neural Network for Accurate Prediction of Quantum Chemistry Properties via Multi-task Learning

“…Development of lower and even linear scaling methods has aroused great interest in the past decade. [9][10][11][12][13][14][15][16] In spite of these advances in the quantum chemical methods, the quick prediction of various electronic structure properties is highly desired in high-throughput searching of large chemical spaces with all possible combinations of functional groups, towards the material or drug design. 4,17,18 Many machine learning methods have been introduced in quantum chemical study for the rapid predictions of atomic forces, molecular energy, and electronic structure properties, which are of great importance for the construction of accurate force fields and complicated potential energy surfaces as well as rational design of various materials and drug-like candidates.…”