Related Experiment Video
Updated: Sep 1, 2025
10:52
Multiscale Sampling of a Heterogeneous Water/Metal Catalyst Interface using Density Functional Theory and Force-Field Molecular Dynamics
Published on: April 12, 2019
12.9K
A Universal Framework for Featurization of Atomistic Systems.
Xiangyun Lei1, Andrew J Medford1
1School of Chemical & Biomolecular Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States.
The Journal of Physical Chemistry Letters
|August 18, 2022
Summary
This study introduces Gaussian multipole (GMP) featurization for machine-learned force fields. GMP offers improved scalability and transferability by using fixed-dimension, element-interpolating features.
Area of Science:
- Computational chemistry
- Materials science
- Machine learning
Background:
- Machine-learning force fields (MLFFs) offer a balance of accuracy and speed.
- Current MLFFs face scalability issues due to element-specific features.
- Developing transferable and efficient MLFFs is crucial for broader applications.
Purpose of the Study:
- Introduce a novel featurization scheme, Gaussian multipole (GMP), for MLFFs.
- Address the limitations of element-specific features in current MLFFs.
- Enhance the scalability and transferability of MLFFs.
Main Methods:
- Developed the Gaussian multipole (GMP) featurization scheme based on atomic electron density multipole expansions.
- Combined GMP with neural networks to create MLFF models.
- Applied and evaluated GMP-based models on benchmark datasets (MD17, QM9) and the OCP dataset.
Main Results:
- GMP featurization yields fixed-dimension feature vectors that interpolate between element types.
- GMP-based models demonstrate high computational efficiency and systematically improvable accuracy.
- Models show reasonable predictive performance on unseen elements and comparable results to graph-convolutional models on the OCP dataset.
Conclusions:
- The GMP featurization scheme effectively addresses the scalability limitations of element-specific features in MLFFs.
- GMP enables the development of more efficient, accurate, and transferable machine-learned force fields.
- This approach fills a critical gap in constructing generalizable MLFFs for diverse chemical systems.
Related Concept Videos
02:45The Quantum-Mechanical Model of an Atom
43.0K
Shortly after de Broglie published his ideas that the electron in a hydrogen atom could be better thought of as being a circular standing wave instead of a particle moving in quantized circular orbits, Erwin Schrödinger extended de Broglie’s work by deriving what is now known as the Schrödinger equation. When Schrödinger applied his equation to hydrogen-like atoms, he was able to reproduce Bohr’s expression for the energy and, thus, the Rydberg formula governing hydrogen spectra.
43.0K
02:00Molecular Models
40.0K
Physical models representing molecular architectures of chemical compounds play essential roles in understanding chemistry. The use of molecular models makes it easier to visualize the structures and shapes of atoms and molecules.
40.0K
01:21Mechanistic Models: Overview of Compartment Models
152
Mechanistic models, a category encompassing both physiological and compartmental modeling, differ from empirical models' approaches to incorporating known factors about the systems being modeled. Empirical models describe data with minimal assumptions, while mechanistic models aim to provide a robust description of available data by specifying assumptions and integrating known factors about the system. Compartmental analysis is a key example of a mechanistic model in pharmacokinetics and...
152
01:17Atomic Structure
12.3K
The Greek philosopher Democritus proposed that everything on Earth is made up of tiny particles called atomos, Greek for "indivisible," from which the modern term "atom" is derived. In the 19th century, John Dalton proposed the atomic theory that is still largely correct today. He put forth five postulates to explain how atoms made up the world around us. (1) All matter is composed of infinitely small particles or atoms. (2) All atoms of a given element are identical to one...
12.3K
01:18First Law: Particles in Two-dimensional Equilibrium
5.2K
Recall that a particle in equilibrium is one for which the external forces are balanced. Static equilibrium involves objects at rest, and dynamic equilibrium involves objects in motion without acceleration; but it is important to remember that these conditions are relative. For instance, an object may be at rest when viewed from one frame of reference, but that same object would appear to be in motion when viewed by someone moving at a constant velocity.
Newton's first law tells us about...
Newton's first law tells us about...
5.2K
02:28Electronic Structure of Atoms
23.7K
An atom comprises protons and neutrons, which are contained inside the dense, central core called the nucleus, with electrons present around the nucleus. Taking into account the wave–particle duality of electrons and the uncertainty in position around the nucleus, quantum mechanics provides a more accurate model for the atomic structure. It describes atomic orbitals as the regions around the nucleus where electrons of discrete energy exist, characterized by four quantum...
23.7K