The MLIP package: moment tensor potentials with MPI and active learning

The subject of this paper is the technology (the "how") of constructing machine-learning interatomic potentials, rather than science (the "what" and "why") of atomistic simulations using machine-learning potentials. Namely, we illustrate how to construct moment tensor potentials using active learning as implemented in the MLIP package, focusing on the efficient ways to sample configurations for the training set, how expanding the training set changes the error of predictions, how to set up ab initio calculations in a cost-effective manner, etc. The MLIP package (short for Machine-Learning Interatomic Potentials) is available at this https URL.

[1]  M. Gastegger,et al.  High-Dimensional Neural Network Potentials for Organic Reactions and an Improved Training Algorithm. , 2015, Journal of chemical theory and computation.

[2]  X. Zhuang,et al.  High thermal conductivity in semiconducting Janus and non-Janus diamanes , 2020, 2006.04091.

[3]  Aidan P Thompson,et al.  Extending the accuracy of the SNAP interatomic potential form. , 2017, The Journal of chemical physics.

[4]  Alexandre Tkatchenko,et al.  Quantum-chemical insights from deep tensor neural networks , 2016, Nature Communications.

[5]  Noam Bernstein,et al.  De novo exploration and self-guided learning of potential-energy surfaces , 2019, npj Computational Materials.

[6]  Ralf Drautz,et al.  Atomic cluster expansion for accurate and transferable interatomic potentials , 2019, Physical Review B.

[7]  Artem R. Oganov,et al.  Machine learning scheme for fast extraction of chemically interpretable interatomic potentials , 2016 .

[8]  Adrian E. Roitberg,et al.  Less is more: sampling chemical space with active learning , 2018, The Journal of chemical physics.

[9]  Christian Trott,et al.  Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials , 2014, J. Comput. Phys..

[10]  Anders S Christensen,et al.  FCHL revisited: Faster and more accurate quantum machine learning. , 2020, The Journal of chemical physics.

[11]  J S Smith,et al.  ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost , 2016, Chemical science.

[12]  P. Korotaev.  Data for "Accessing thermal conductivity of complex compounds by machine learning interatomic potentials" , 2019 .

[13]  Andrea Grisafi,et al.  Symmetry-Adapted Machine Learning for Tensorial Properties of Atomistic Systems. , 2017, Physical review letters.

[14]  E Weinan,et al.  Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics , 2017, Physical review letters.

[15]  Georg Kresse,et al.  On-the-fly machine learning force field generation: Application to melting points , 2019, Physical Review B.

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

[17]  A. Shapeev,et al.  Ring polymer molecular dynamics and active learning of moment tensor potential for gas-phase barrierless reactions: Application to S + H2. , 2019, The Journal of chemical physics.

[18]  Gábor Csányi,et al.  Comparing molecules and solids across structural and alchemical space. , 2015, Physical chemistry chemical physics : PCCP.

[19]  Roberto Car,et al.  A Novel Scheme for Accurate Md Simulations of Large Systems , 1997 .

[20]  Samuel S. Schoenholz,et al.  Fast machine learning models of electronic and energetic properties consistently reach approximation errors better than DFT accuracy , 2017 .

[21]  R. Batra,et al.  Physically informed artificial neural networks for atomistic modeling of materials , 2018, Nature Communications.

[22]  O. A. von Lilienfeld,et al.  Communication: Understanding molecular representations in machine learning: The role of uniqueness and target similarity. , 2016, The Journal of chemical physics.

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

[24]  Nikolaus Hansen,et al.  USPEX - Evolutionary crystal structure prediction , 2006, Comput. Phys. Commun..

[25]  G. Kresse,et al.  Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set , 1996 .

[26]  Christoph Ortner,et al.  Incompleteness of Atomic Structure Representations. , 2020, Physical review letters.

[27]  Cas van der Oord,et al.  Regularised atomic body-ordered permutation-invariant polynomials for the construction of interatomic potentials , 2019, Mach. Learn. Sci. Technol..

[28]  Klaus-Robert Müller,et al.  Finding Density Functionals with Machine Learning , 2011, Physical review letters.

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

[30]  S. Goreinov,et al.  How to find a good submatrix , 2010 .

[31]  Conrad W. Rosenbrock,et al.  Machine-learned multi-system surrogate models for materials prediction , 2018, npj Computational Materials.

[32]  T. Rabczuk,et al.  Nanoporous C3N4, C3N5 and C3N6 nanosheets; novel strong semiconductors with low thermal conductivities and appealing optical/electronic properties , 2020, 2006.03889.

[33]  Sergei Manzhos,et al.  Neural network‐based approaches for building high dimensional and quantum dynamics‐friendly potential energy surfaces , 2015 .

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

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

[36]  A. Shapeev,et al.  Lattice dynamics simulation using machine learning interatomic potentials , 2020 .

[37]  A. Shapeev,et al.  Automated calculation of thermal rate coefficients using ring polymer molecular dynamics and machine-learning interatomic potentials with active learning. , 2018, Physical chemistry chemical physics : PCCP.

[38]  Anubhav Jain,et al.  The Materials Application Programming Interface (API): A simple, flexible and efficient API for materials data based on REpresentational State Transfer (REST) principles , 2015 .

[39]  Michele Parrinello,et al.  Generalized neural-network representation of high-dimensional potential-energy surfaces. , 2007, Physical review letters.

[40]  Andre Lomaka,et al.  Linearization of moment tensor potentials for multicomponent systems with a preliminary assessment for short-range interaction energy in water dimer and trimer. , 2020, The Journal of chemical physics.

[41]  A. Shapeev.  Accurate representation of formation energies of crystalline alloys with many components , 2016, 1612.03359.

[42]  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.

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

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

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

[46]  Justin S. Smith,et al.  Hierarchical modeling of molecular energies using a deep neural network. , 2017, The Journal of chemical physics.

[47]  A. Thompson,et al.  Explicit Multi-element Extension of the Spectral Neighbor Analysis Potential for Chemically Complex Systems. , 2020, The journal of physical chemistry. A.

[48]  Alexander V. Shapeev,et al.  Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials , 2015, Multiscale Model. Simul..

[49]  K. Khoo,et al.  Applying a machine learning interatomic potential to unravel the effects of local lattice distortion on the elastic properties of multi-principal element alloys , 2019, Journal of Alloys and Compounds.

[50]  A. Shapeev,et al.  Improving accuracy of interatomic potentials: more physics or more data? A case study of silica , 2018, Materials Today Communications.

[51]  E Weinan,et al.  Active Learning of Uniformly Accurate Inter-atomic Potentials for Materials Simulation , 2018, Physical Review Materials.

[52]  Álvaro Vázquez-Mayagoitia,et al.  Machine Learning Inter-Atomic Potentials Generation Driven by Active Learning: A Case Study for Amorphous and Liquid Hafnium dioxide , 2019, ArXiv.

[53]  T. Rabczuk,et al.  Efficient machine-learning based interatomic potentialsfor exploring thermal conductivity in two-dimensional materials , 2020, Journal of Physics: Materials.

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

[55]  Klaus-Robert Müller,et al.  SchNet: A continuous-filter convolutional neural network for modeling quantum interactions , 2017, NIPS.

[56]  Nongnuch Artrith,et al.  High-dimensional neural network potentials for metal surfaces: A prototype study for copper , 2012 .

[57]  G. Kresse,et al.  Ab initio molecular dynamics for liquid metals. , 1993 .

[58]  Y. Ikeda,et al.  Ab initio vibrational free energies including anharmonicity for multicomponent alloys , 2019, npj Computational Materials.

[59]  Burr Settles,et al.  Active Learning , 2012, Synthesis Lectures on Artificial Intelligence and Machine Learning.

[60]  S. Curtarolo,et al.  AFLOW: An automatic framework for high-throughput materials discovery , 2012, 1308.5715.

[61]  Jonathan Vandermause,et al.  On-the-fly active learning of interpretable Bayesian force fields for atomistic rare events , 2020, npj Computational Materials.

[62]  T. Rabczuk,et al.  Exploring phononic properties of two-dimensional materials using machine learning interatomic potentials , 2020, 2005.04913.

[63]  A. Ruban,et al.  Performance of the standard exchange-correlation functionals in predicting melting properties fully from first principles: Application to Al and magnetic Ni , 2020, Physical Review B.

[64]  A. Shapeev,et al.  Accessing thermal conductivity of complex compounds by machine learning interatomic potentials , 2019, Physical Review B.

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

[66]  Konstantin Gubaev,et al.  Machine learning of molecular properties: Locality and active learning. , 2017, The Journal of chemical physics.

[67]  Technology,et al.  Moment tensor potentials as a promising tool to study diffusion processes , 2018, Computational Materials Science.

[68]  M C Payne,et al.  "Learn on the fly": a hybrid classical and quantum-mechanical molecular dynamics simulation. , 2004, Physical review letters.

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

[70]  Alexander V. Shapeev,et al.  Accelerating crystal structure prediction by machine-learning interatomic potentials with active learning , 2018, Physical Review B.

[71]  Gábor Csányi,et al.  Accuracy and transferability of Gaussian approximation potential models for tungsten , 2014 .

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

[73]  Adarsh Balasubramanian,et al.  Fast, accurate, and transferable many-body interatomic potentials by symbolic regression , 2019, npj Computational Materials.

[74]  Timon Rabczuk,et al.  Machine-learning interatomic potentials enable first-principles multiscale modeling of lattice thermal conductivity in graphene/borophene heterostructures , 2020, Materials Horizons.

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

[76]  Klaus-Robert Müller,et al.  Exploring chemical compound space with quantum-based machine learning , 2020, Nature Reviews Chemistry.

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

[78]  Samuel S. Schoenholz,et al.  Neural Message Passing for Quantum Chemistry , 2017, ICML.