Training machine learning potentials for reactive systems: A Colab tutorial on basic models

Abstract

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 (https://cc-ats.github.io/mlp_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 Müller‐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.

Generating High-Precision Force Fields for Molecular Dynamics Simulations to Study Chemical Reaction Mechanisms Using Molecular Configuration Transformer

“…With the rapid development of artificial intelligence (AI), the application of machine-learning-based force fields has become popular in recent years. In particular, the recent development of those deep neural network-based force fields, such as BPNN, DPMD, EANN, SchNet, and PhysNet, is expected to help achieve both high speed and accuracy. , …”

“…In particular, the recent development of those deep neural network-based force fields, such as BPNN, 56 DPMD, 57 EANN, 58 SchNet, 59 and PhysNet, 60 is expected to help achieve both high speed and accuracy. 61,62 Here, we present a strategy for generating high-precision force fields for chemical reactions. This force field is based on the molecular configuration transformer (MolCT) model, a graph neural network (GNN)-based deep molecular model developed in our group.…”