Parallel Implementation of Nonadditive Gaussian Process Potentials for Monte Carlo Simulations

A strategy is presented to implement Gaussian process potentials in molecular simulations through parallel programming. Attention is focused on the three-body nonadditive energy, though all algorithms extend straightforwardly to the additive energy. The method to distribute pairs and triplets between processes is general to all potentials. Results are presented for a simulation box of argon, including full box and atom displacement calculations, which are relevant to Monte Carlo simulation. Data on speed-up are presented for up to 120 processes across four nodes. A 4-fold speed-up is observed over five processes, extending to 20-fold over 40 processes and 30-fold over 120 processes.

[1]  J. Bowman,et al.  Permutationally invariant polynomial regression for energies and gradients, using reverse differentiation, achieves orders of magnitude speed-up with high precision compared to other machine learning methods. , 2021, The Journal of chemical physics.

[2]  Yaolong Zhang,et al.  REANN: A PyTorch-based end-to-end multi-functional deep neural network package for molecular, reactive, and periodic systems. , 2021, The Journal of chemical physics.

[3]  Richard S. Graham,et al.  Gaussian process models of potential energy surfaces with boundary optimization. , 2021, The Journal of chemical physics.

[4]  R. Sadus,et al.  Interatomic Interactions Responsible for the Solid-Liquid and Vapor-Liquid Phase Equilibria of Neon. , 2021, The journal of physical chemistry. B.

[5]  Klaus-Robert Müller,et al.  SpookyNet: Learning force fields with electronic degrees of freedom and nonlocal effects , 2021, Nature Communications.

[6]  Ulf R. Pedersen,et al.  Solid-liquid coexistence of neon, argon, krypton, and xenon studied by simulations. , 2020, The Journal of chemical physics.

[7]  Gábor Csányi,et al.  Machine learning interatomic potential developed for molecular simulations on thermal properties of β-Ga2O3. , 2020, The Journal of chemical physics.

[8]  R. Krems,et al.  Gaussian process model of 51-dimensional potential energy surface for protonated imidazole dimer. , 2020, The Journal of chemical physics.

[9]  M. Sansom,et al.  Water in Nanopores and Biological Channels: A Molecular Simulation Perspective , 2020, Chemical reviews.

[10]  Mohamed Ali Boussaidi,et al.  Random Sampling High Dimensional Model Representation Gaussian Process Regression (RS-HDMR-GPR) for Multivariate Function Representation: Application to Molecular Potential Energy Surfaces. , 2020, The journal of physical chemistry. A.

[11]  Ove Christiansen,et al.  A Gaussian process regression adaptive density guided approach for potential energy surface construction. , 2020, The Journal of chemical physics.

[12]  Gábor Csányi,et al.  Machine-learned interatomic potentials by active learning: amorphous and liquid hafnium dioxide , 2020, npj Computational Materials.

[13]  Matthew J Burn,et al.  Creating Gaussian process regression models for molecular simulations using adaptive sampling. , 2020, The Journal of chemical physics.

[14]  Tim Mueller,et al.  Machine learning for interatomic potential models. , 2020, The Journal of chemical physics.

[15]  R. Krems,et al.  Interpolation and extrapolation of global potential energy surfaces for polyatomic systems by Gaussian processes with composite kernels. , 2019, Journal of chemical theory and computation.

[16]  R. Sadus,et al.  Fully a priori prediction of the vapor-liquid equilibria of Ar, Kr, and Xe from ab initio two-body plus three-body interatomic potentials. , 2019, The Journal of chemical physics.

[17]  J. Behler,et al.  A Performance and Cost Assessment of Machine Learning Interatomic Potentials. , 2019, The journal of physical chemistry. A.

[18]  Elena Uteva,et al.  Active learning in Gaussian process interpolation of potential energy surfaces. , 2018, The Journal of chemical physics.

[19]  Aditya Kamath,et al.  Neural networks vs Gaussian process regression for representing potential energy surfaces: A comparative study of fit quality and vibrational spectrum accuracy. , 2018, The Journal of chemical physics.

[20]  Mark S. Gordon,et al.  Perspective: Ab initio force field methods derived from quantum mechanics , 2018 .

[21]  Nicola Gaston,et al.  Building machine learning force fields for nanoclusters. , 2018, The Journal of chemical physics.

[22]  Richard D Wilkinson,et al.  Interpolation of intermolecular potentials using Gaussian processes. , 2017, The Journal of chemical physics.

[23]  J. Behler.  Perspective: Machine learning potentials for atomistic simulations. , 2016, The Journal of chemical physics.

[24]  T. Ilyina.  Climate science: Hidden trends in the ocean carbon sink , 2016, Nature.

[25]  Jae-Young Choi,et al.  Selective Gas Transport Through Few-Layered Graphene and Graphene Oxide Membranes , 2013, Science.

[26]  Joshua D. Knowles,et al.  Accuracy and tractability of a kriging model of intramolecular polarizable multipolar electrostatics and its application to histidine , 2013, J. Comput. Chem..

[27]  Paul L. A. Popelier,et al.  Polarisable multipolar electrostatics from the machine learning method Kriging: an application to alanine , 2012, Theoretical Chemistry Accounts.

[28]  P. Popelier,et al.  Intramolecular polarisable multipolar electrostatics from the machine learning method Kriging , 2011 .

[29]  Douglas B Kell,et al.  Optimal construction of a fast and accurate polarisable water potential based on multipole moments trained by machine learning. , 2009, Physical chemistry chemical physics : PCCP.

[30]  Richard J. Beckman,et al.  A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code , 2000, Technometrics.

[31]  Wei-Liem Loh.  On Latin hypercube sampling , 1996 .

[32]  M. Stein.  Large sample properties of simulations using latin hypercube sampling , 1987 .

[33]  C. Brooks.  Computer simulation of liquids , 1989 .