Machine Learning Force Fields: Construction, Validation, and Outlook

Force fields developed with machine learning methods in tandem with quantum mechanics are beginning to find merit, given their (i) low cost, (ii) accuracy, and (iii) versatility. Recently, we proposed one such approach, wherein, the vectorial force on an atom is computed directly from its environment. Here, we discuss the multistep workflow required for their construction, which begins with generating diverse reference atomic environments and force data, choosing a numerical representation for the atomic environments, down selecting a representative training set, and lastly the learning method itself, for the case of Al. The constructed force field is then validated by simulating complex materials phenomena such as surface melting and stress–strain behavior, that truly go beyond the realm of ab initio methods, both in length and time scales. To make such force fields truly versatile an attempt to estimate the uncertainty in force predictions is put forth, allowing one to identify areas of poor performance...

[1]  J. Behler.  Neural network potential-energy surfaces in chemistry: a tool for large-scale simulations. , 2011, Physical chemistry chemical physics : PCCP.

[2]  Steve Plimpton,et al.  Fast parallel algorithms for short-range molecular dynamics , 1993 .

[3]  Michael Rabadi,et al.  Kernel Methods for Machine Learning , 2015 .

[4]  A. Bogaerts,et al.  Numerical Study of the Size-Dependent Melting Mechanisms of Nickel Nanoclusters , 2009 .

[5]  Bernhard Schölkopf,et al.  Kernel Principal Component Analysis , 1997, ICANN.

[6]  J. Behler.  Atom-centered symmetry functions for constructing high-dimensional neural network potentials. , 2011, The Journal of chemical physics.

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

[8]  J. Elliott.  Novel approaches to multiscale modelling in materials science , 2011 .

[9]  Blöchl,et al.  Projector augmented-wave method. , 1994, Physical review. B, Condensed matter.

[10]  A. Gross,et al.  Representing high-dimensional potential-energy surfaces for reactions at surfaces by neural networks , 2004 .

[11]  J. Kermode,et al.  Modelling defects in Ni–Al with EAM and DFT calculations , 2016 .

[12]  Kresse,et al.  Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. , 1996, Physical review. B, Condensed matter.

[13]  C. Freeman,et al.  Use of Force Fields in Materials Modeling , 2007 .

[14]  S. Alavi,et al.  Molecular dynamics simulations of the melting of aluminum nanoparticles. , 2006, The journal of physical chemistry. A.

[15]  Car,et al.  Unified approach for molecular dynamics and density-functional theory. , 1985, Physical review letters.

[16]  Anubhav Jain,et al.  From the computer to the laboratory: materials discovery and design using first-principles calculations , 2012, Journal of Materials Science.

[17]  K. Burke.  Perspective on density functional theory. , 2012, The Journal of chemical physics.

[18]  Jacobsen,et al.  Interatomic interactions in the effective-medium theory. , 1987, Physical review. B, Condensed matter.

[19]  Matthias Rupp,et al.  Machine learning for quantum mechanics in a nutshell , 2015 .

[20]  Klaus-Robert Müller,et al.  Assessment and Validation of Machine Learning Methods for Predicting Molecular Atomization Energies. , 2013, Journal of chemical theory and computation.

[21]  Rampi Ramprasad,et al.  Adaptive machine learning framework to accelerate ab initio molecular dynamics , 2015 .

[22]  R. Arróyave,et al.  Finite-temperature elasticity of fcc Al: Atomistic simulations and ultrasonic measurements , 2011 .

[23]  M. Karplus,et al.  The distance fluctuation criterion for melting: Comparison of square-well and Morse potential models for clusters and homopolymers , 2002 .

[24]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[25]  Marcel Abendroth,et al.  Data Mining Practical Machine Learning Tools And Techniques With Java Implementations , 2016 .

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

[27]  J. Zhong,et al.  Melting and premelting of carbon nanotubes , 2007 .

[28]  Gunnar Rätsch,et al.  An introduction to kernel-based learning algorithms , 2001, IEEE Trans. Neural Networks.

[29]  Zhenwei Li,et al.  Molecular dynamics with on-the-fly machine learning of quantum-mechanical forces. , 2015, Physical review letters.

[30]  Wang,et al.  Generalized gradient approximation for the exchange-correlation hole of a many-electron system. , 1996, Physical review. B, Condensed matter.

[31]  John R. Kitchin,et al.  Neural network and ReaxFF comparison for Au properties , 2016 .

[32]  Louis G. Birta,et al.  Modelling and Simulation , 2013, Simulation Foundations, Methods and Applications.

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

[34]  P. Groenen,et al.  Modern multidimensional scaling , 1996 .

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

[36]  Jörg Neugebauer,et al.  Density functional theory in materials science , 2013, Wiley interdisciplinary reviews. Computational molecular science.

[37]  Landman,et al.  Disordering and melting of aluminum surfaces. , 1988, Physical review letters.

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

[39]  Ellad B. Tadmor,et al.  Modeling Materials: Continuum, Atomistic and Multiscale Techniques , 2011 .

[40]  Rampi Ramprasad,et al.  Learning scheme to predict atomic forces and accelerate materials simulations , 2015, 1505.02701.