Machine learning with bond information for local structure optimizations in surface science.

Local optimization of adsorption systems inherently involves different scales: within the substrate, within the molecule, and between the molecule and the substrate. In this work, we show how the explicit modeling of different characteristics of the bonds in these systems improves the performance of machine learning methods for optimization. We introduce an anisotropic kernel in the Gaussian process regression framework that guides the search for the local minimum, and we show its overall good performance across different types of atomic systems. The method shows a speed-up of up to a factor of two compared with the fastest standard optimization methods on adsorption systems. Additionally, we show that a limited memory approach is not only beneficial in terms of overall computational resources but can also result in a further reduction of energy and force calculations.

[1]  Henrik Lund Mortensen,et al.  Atomistic Structure Learning Algorithm with surrogate energy model relaxation , 2020, ArXiv.

[2]  Andrew Gordon Wilson,et al.  GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration , 2018, NeurIPS.

[3]  Alexander Denzel,et al.  A Hessian Matrix Update Scheme for Transition-State Search based on Gaussian Process Regression. , 2020, Journal of chemical theory and computation.

[4]  Noam Bernstein,et al.  A universal preconditioner for simulating condensed phase materials. , 2016, The Journal of chemical physics.

[5]  Aki Vehtari,et al.  Nudged elastic band calculations accelerated with Gaussian process regression based on inverse inter-atomic distances. , 2019, Journal of chemical theory and computation.

[6]  Matthias Poloczek,et al.  Scalable Global Optimization via Local Bayesian Optimization , 2019, NeurIPS.

[7]  Filippo Federici Canova,et al.  DScribe: Library of Descriptors for Machine Learning in Materials Science , 2019, Comput. Phys. Commun..

[8]  Ralf Meyer,et al.  Geometry optimization using Gaussian process regression in internal coordinate systems. , 2020, The Journal of chemical physics.

[9]  R. Kondor,et al.  Gaussian approximation potentials: the accuracy of quantum mechanics, without the electrons. , 2009, Physical review letters.

[10]  Volker L. Deringer,et al.  Machine learning based interatomic potential for amorphous carbon , 2016, 1611.03277.

[11]  Karsten Wedel Jacobsen,et al.  Local Bayesian optimizer for atomic structures , 2018, Physical Review B.

[12]  Volker L. Deringer,et al.  Data-Driven Learning of Total and Local Energies in Elemental Boron. , 2017, Physical review letters.

[13]  Christoph Ortner,et al.  A preconditioning scheme for minimum energy path finding methods. , 2019, The Journal of chemical physics.

[14]  Paul W. Ayers,et al.  Quasi-Newton parallel geometry optimization methods. , 2010, The Journal of chemical physics.

[15]  N. A. Romero,et al.  Electronic structure calculations with GPAW: a real-space implementation of the projector augmented-wave method , 2010, Journal of physics. Condensed matter : an Institute of Physics journal.

[16]  J. Nørskov,et al.  Improved adsorption energetics within density-functional theory using revised Perdew-Burke-Ernzerhof functionals , 1999 .

[17]  W. Kohn,et al.  Self-Consistent Equations Including Exchange and Correlation Effects , 1965 .

[18]  R. Kondor,et al.  On representing chemical environments , 2012, 1209.3140.

[19]  Beatriz Cordero,et al.  Covalent radii revisited. , 2008, Dalton transactions.

[20]  Klaus-Robert Müller,et al.  Machine learning of accurate energy-conserving molecular force fields , 2016, Science Advances.

[21]  Stephen Tyree,et al.  Exact Gaussian Processes on a Million Data Points , 2019, NeurIPS.

[22]  Santiago Alvarez,et al.  A cartography of the van der Waals territories. , 2013, Dalton transactions.

[23]  Eric Jones,et al.  SciPy: Open Source Scientific Tools for Python , 2001 .

[24]  Pekka Koskinen,et al.  Structural relaxation made simple. , 2006, Physical review letters.

[25]  Roland Lindh,et al.  ON THE USE OF A HESSIAN MODEL FUNCTION IN MOLECULAR GEOMETRY OPTIMIZATIONS , 1995 .

[26]  Karsten W Jacobsen,et al.  Exploration versus Exploitation in Global Atomistic Structure Optimization. , 2018, The journal of physical chemistry. A.

[27]  Andrew A Peterson,et al.  Acceleration of saddle-point searches with machine learning. , 2016, The Journal of chemical physics.

[28]  Alexander Denzel,et al.  Gaussian Process Regression for Geometry Optimization , 2018, 2009.05803.

[29]  L. Curtiss,et al.  Assessment of Gaussian-2 and density functional theories for the computation of enthalpies of formation , 1997 .

[30]  D K Smith,et al.  Numerical Optimization , 2001, J. Oper. Res. Soc..

[31]  Christoph Ortner,et al.  Preconditioners for the geometry optimisation and saddle point search of molecular systems , 2018, Scientific Reports.

[32]  O. Anatole von Lilienfeld,et al.  On the role of gradients for machine learning of molecular energies and forces , 2020, Mach. Learn. Sci. Technol..

[33]  Alexander Denzel,et al.  Gaussian Process Regression for Transition State Search. , 2018, Journal of chemical theory and computation.

[34]  James P. Sethna,et al.  The potential of atomistic simulations and the knowledgebase of interatomic models , 2011 .

[35]  Adam S. Foster,et al.  Machine learning hydrogen adsorption on nanoclusters through structural descriptors , 2018, npj Computational Materials.

[36]  Aki Vehtari,et al.  Nudged elastic band calculations accelerated with Gaussian process regression. , 2017, The Journal of chemical physics.

[37]  Michael Walter,et al.  The atomic simulation environment-a Python library for working with atoms. , 2017, Journal of physics. Condensed matter : an Institute of Physics journal.

[38]  Gus L. W. Hart,et al.  Accelerating high-throughput searches for new alloys with active learning of interatomic potentials , 2018, Computational Materials Science.

[39]  Takashi Miyake,et al.  Crystal structure prediction accelerated by Bayesian optimization , 2018 .

[40]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[41]  Matthias Poloczek,et al.  Bayesian Optimization with Gradients , 2017, NIPS.

[42]  Ove Christiansen,et al.  Gaussian process regression to accelerate geometry optimizations relying on numerical differentiation. , 2018, The Journal of chemical physics.

[43]  K. Müller,et al.  Fast and accurate modeling of molecular atomization energies with machine learning. , 2011, Physical review letters.

[44]  Fernando Nogueira,et al.  A Tutorial on Density Functional Theory , 2003 .

[45]  Jorge Nocedal,et al.  A Limited Memory Algorithm for Bound Constrained Optimization , 1995, SIAM J. Sci. Comput..

[46]  Jukka Corander,et al.  Bayesian inference of atomistic structure in functional materials , 2017, npj Computational Materials.

[47]  Thomas Bligaard,et al.  Low-Scaling Algorithm for Nudged Elastic Band Calculations Using a Surrogate Machine Learning Model. , 2018, Physical review letters.

[48]  P. Hohenberg,et al.  Inhomogeneous Electron Gas , 1964 .

[49]  Alexander V. Shapeev,et al.  Active learning of linearly parametrized interatomic potentials , 2016, 1611.09346.

[50]  Milica Todorovic,et al.  Efficient Cysteine Conformer Search with Bayesian Optimization , 2020, 2006.15006.

[51]  Malthe K. Bisbo,et al.  Efficient Global Structure Optimization with a Machine-Learned Surrogate Model. , 2019, Physical review letters.

[52]  K. Müller,et al.  Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space , 2015, The journal of physical chemistry letters.

[53]  G. Kresse,et al.  From ultrasoft pseudopotentials to the projector augmented-wave method , 1999 .