.png){kind=icon}
Return
Training machine learning potentials for reactive systems: A Colab tutorial on basic models
.png){kind=icon}
PRE
AI
DOI：10.1002/jcc.27269.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In the last several years, there has been a surge in the development of machine learning potential (MLP) models for describing molecular systems. We are interested in a particular area of this field - the training of system-specific MLPs for reactive systems - with the goal of using these MLPs to accelerate free energy simulations of chemical and enzyme reactions. To help new members in our labs become familiar with the basic techniques, we have put together a self-guided Colab tutorial (), which we expect to be also useful to other young researchers in the community. Our tutorial begins with the introduction of simple feedforward neural network (FNN) and kernel-based (using Gaussian process regression, GPR) models by fitting the two-dimensional Muller-Brown potential. Subsequently, two simple descriptors are presented for extracting features of molecular systems: symmetry functions (including the ANI variant) and embedding neural networks (such as DeepPot-SE). Lastly, these features will be fed into FNN and GPR models to reproduce the energies and forces for the molecular configurations in a Claisen rearrangement reaction. A self-guided Colab tutorial about machine learning potential for reactive systems are presented in this work. The tutorial begins with the introduction of feedforward neural network and kernel-based models by fitting the two-dimensional Muller-Brown potential. Subsequently, two simple descriptors are presented for extracting features of molecular systems: symmetry functions and embedding neural networks. Lastly, these features will be used to reproduce the energies and forces for the molecular configurations in a Claisen rearrangement reaction.image
Keywords:
Gaussian process regression
machine learning potential
neural network
tutorial
Journal
IF:
4.8
Papers:
7.1K
Citations:
6.1W
Organization
View more organizations
Cited Papers
ShareSaveShareSaveShareSaveShareSaveE(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials.png)
NATURE COMMUNICATIONS
OAAI
NATURE COMMUNICATIONSIF15.7
OAAIShareSaveMachine learning enables long time scale molecular photodynamics simulationsChemical ScienceOAAI
Chemical ScienceIF7.4
OAAIShareSaveA simplified charge projection scheme for long-range electrostatics in ab initio QM/MM calculationsOAAI
OAAIShareSaveShareSave View more