Gaussian approximation potentials: A brief tutorial introduction

2015/04/27 by Albert P. Bartók, Gábor Csányi, Gábor Cśanyi · 615 citations
Biochemistry, Genetics and Molecular Biology · Computer Science · Materials Science · #Artificial intelligence #Computational Drug Discovery Methods #Computational science #Computer science #Gaussian #Machine Learning in Materials Science #Physics #Protein Structure and Dynamics #Quantum mechanics #Sandbox (software development) #Software #Statistical physics #Theoretical computer science #Variety (cybernetics) #Work (physics)

paper · pdf · doi:10.1002/qua.24927

published in International Journal of Quantum Chemistry 115(16), 1051-1057 (Wiley)

openalex publication_date 2015/04/27 · openalex created_date 2016/06/24 · openalex updated_date 2026/08/05

Abstract

We present a swift walk‐through of our recent work that uses machine learning to fit interatomic potentials based on quantum mechanical data. We describe our Gaussian approximation potentials (GAP) framework, discuss a variety of descriptors, how to train the model on total energies and derivatives, and the simultaneous use of multiple models of different complexity. We also show a small example using QUIP, the software sandbox implementation of GAP that is available for noncommercial use. © 2015 Wiley Periodicals, Inc.

Cited by

• Machine learning for nonadiabatic molecular dynamics: best practices and recent progress
• Neural Network Potentials: A Concise Overview of Methods
• Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer
• Measuring transferability issues in machine-learning force fields: The\n example of Gold-Iron interactions with linearized potentials
• Heat transfer enhancement of N-Ga-Al semiconductors heterogeneous interfaces
• Fast machine learned interatomic potential for hydrogen-induced embrittlement in α-Fe
• A Review of AI-Driven Approaches for Nanoscale Heat Conduction and Radiation
• Fast and Accurate Uncertainty Estimation in Chemical Machine Learning
• Four Generations of High-Dimensional Neural Network Potentials
• Gaussian Process Regression for Materials and Molecules
• A perspective marking 20 years of using permutationally invariant polynomials for molecular potentials
• Atomic-Scale Origins of Oxidation Resistance in Amorphous Boron Nitride
• Machine Learning and Materials Informatics: Recent Applications and Prospects
• Thermal Conductivity Limits of MoS₂ and MoSe₂: Revisiting High-Order Anharmonic Lattice Dynamics with Machine Learning Potentials
• Effects of training machine-learning potentials for radiation damage simulations using different pseudopotentials
• IR Spectroscopic Studies of Gas-Phase Peptides
• Hybrid Quantum--Classical Machine Learning Potential with Variational Quantum Circuits
• Training Data Set Refinement for the Machine Learning Potential of Li-Si Alloys via Structural Similarity Analysis
• Fluid flow inside slit-shaped nanopores: The role of surface morphology at the molecular scale
• Hydrogen Diffusion in Garnet: Insights From Atomistic Simulations
• Prediction of Atomization Energies of Au13+ Clusters through the Machine Learning Approach
• Clarifying the Ti-V Phase Diagram Using First-Principles Calculations and Bayesian Learning
• Particle Swarm Based Hyper-Parameter Optimization for Machine Learned\n Interatomic Potentials
• Machine-Learned Potentials for Solvation Modeling
• Machine Learning Interatomic Potentials: library for efficient training, model development and simulation of molecular systems
• MT-CGCNN: Integrating Crystal Graph Convolutional Neural Network with Multitask Learning for Material Property Prediction
• Self-consistency error correction for accurate machine learning potentials from variational Monte Carlo
• Uncovering structure-property relationships of materials by subgroup discovery
• Deep machine learning potentials for multicomponent metallic melts: development, predictability and compositional transferability
• Molecular force fields with gradient-domain machine learning (GDML): Comparison and synergies with classical force fields
• Hierarchical modeling of molecular energies using a deep neural network
• High-entropy ceramics: Review of principles, production and applications
• Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights
• Structure-based sampling and self-correcting machine learning for accurate calculations of potential energy surfaces and vibrational levels
• Bayesian force fields from active learning for simulation of inter-dimensional transformation of stanene
• Nudged elastic band calculations accelerated with Gaussian process regression
• Bayesian optimization for the inverse scattering problem in quantum reaction dynamics
• Dynamically polarisable force-fields for surface simulations via\n multi-output classification Neural Networks
• Gaussian processes for choosing laser parameters for driven, dissipative Rydberg aggregates
• Neural network potential from bispectrum components: A case study on crystalline silicon
• A Hessian-based assessment of atomic forces for training machine learning interatomic potentials
• Modeling Molecular Interactions in Water: From Pairwise to Many-Body Potential Energy Functions. [europepmc]
• Nonadiabatic Excited-State Dynamics with Machine Learning. [europepmc]
• Towards exact molecular dynamics simulations with machine-learned force fields. [europepmc]
• Reactivity of Amorphous Carbon Surfaces: Rationalizing the Role of Structural Motifs in Functionalization Using Machine Learning. [europepmc]
• Revealing the intrinsic nature of the mid-gap defects in amorphous Ge 2 Sb 2 Te 5 . [europepmc]
• Efficient Generation of Permutationally Invariant Potential Energy Surfaces for Large Molecules. [europepmc]
• Kernel-Based Machine Learning for Efficient Simulations of Molecular Liquids. [europepmc]
• Big-Data Science in Porous Materials: Materials Genomics and Machine Learning. [europepmc]
• Quantum chemical accuracy from density functional approximations via machine learning. [europepmc]
• Dynamical strengthening of covalent and non-covalent molecular interactions by nuclear quantum effects at finite temperature. [europepmc]
• Machine Learning Force Fields. [europepmc]
• MLatom 2: An Integrative Platform for Atomistic Machine Learning. [europepmc]
• Data-efficient machine learning for molecular crystal structure prediction. [europepmc]
• Gaussian Process Regression for Materials and Molecules. [europepmc]
• A transferable active-learning strategy for reactive molecular force fields. [europepmc]
• A Differentiable Neural-Network Force Field for Ionic Liquids. [europepmc]
• Machine learning potential era of zeolite simulation. [europepmc]
• A Hybrid Machine Learning Approach for Structure Stability Prediction in Molecular Co-crystal Screenings. [europepmc]
• Accurate Computational Prediction of Core-Electron Binding Energies in Carbon-Based Materials: A Machine-Learning Model Combining Density-Functional Theory and GW . [europepmc]
• Long-range dispersion-inclusive machine learning potentials for structure search and optimization of hybrid organic-inorganic interfaces. [europepmc]
• Efficient interatomic descriptors for accurate machine learning force fields of extended molecules. [europepmc]
• Uncertainty-driven dynamics for active learning of interatomic potentials. [europepmc]
• Exploring the frontiers of condensed-phase chemistry with a general reactive machine learning potential. [europepmc]
• Hyperactive learning for data-driven interatomic potentials. [europepmc]
• Data Generation for Machine Learning Interatomic Potentials and Beyond. [europepmc]

Related

• Gaussian Approximation Potentials: a brief tutorial introduction
• Gaussian Approximation Potentials: theory, software implementation and application examples
• Gaussian Approximation Potentials: the accuracy of quantum mechanics, without the electrons
• Gaussian Approximation Potential: an interatomic potential derived from first principles Quantum Mechanics
• The transformative capability of quantum-accurate machine learning interatomic potentials