A Universal 3D Voxel Descriptor for Solid-State Material Informatics with Deep Convolutional Neural Networks

Material informatics (MI) is a promising approach to liberate us from the time-consuming Edisonian (trial and error) process for material discoveries, driven by machine-learning algorithms. Several descriptors, which are encoded material features to feed computers, were proposed in the last few decades. Especially to solid systems, however, their insufficient representations of three dimensionality of field quantities such as electron distributions and local potentials have critically hindered broad and practical successes of the solid-state MI. We develop a simple, generic 3D voxel descriptor that compacts any field quantities, in such a suitable way to implement convolutional neural networks (CNNs). We examine the 3D voxel descriptor encoded from the electron distribution by a regression test with 680 oxides data. The present scheme outperforms other existing descriptors in the prediction of Hartree energies that are significantly relevant to the long-wavelength distribution of the valence electrons. The results indicate that this scheme can forecast any functionals of field quantities just by learning sufficient amount of data, if there is an explicit correlation between the target properties and field quantities. This 3D descriptor opens a way to import prominent CNNs-based algorithms of supervised, semi-supervised and reinforcement learnings into the solid-state MI.

[1]  Van Vechten,et al.  Quantum Dielectric Theory of Electronegativity in Covalent Systems. I. Electronic Dielectric Constant , 1969 .

[2]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[3]  Andrew L. Maas.  Rectifier Nonlinearities Improve Neural Network Acoustic Models , 2013 .

[4]  G. B. Olson,et al.  Computational Design of Hierarchically Structured Materials , 1997 .

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

[6]  Felix A Faber,et al.  Machine Learning Energies of 2 Million Elpasolite (ABC_{2}D_{6}) Crystals. , 2015, Physical review letters.

[7]  Klaus H. Maier-Hein,et al.  Deep MRI brain extraction: A 3D convolutional neural network for skull stripping , 2016, NeuroImage.

[8]  Shuichi Iwata,et al.  Data-Driven Atomic Environment Prediction for Binaries Using the Mendeleev Number. Part 1. Composition AB. , 2004 .

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

[10]  Burke,et al.  Generalized Gradient Approximation Made Simple. , 1996, Physical review letters.

[11]  Ming Yang,et al.  3D Convolutional Neural Networks for Human Action Recognition , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Alex Zunger,et al.  Systematization of the stable crystal structure of all AB-type binary compounds: A pseudopotential orbital-radii approach , 1980 .

[13]  M. Rupp,et al.  Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties , 2013, 1307.2918.

[14]  R. Parr.  Density-functional theory of atoms and molecules , 1989 .

[15]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[16]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[17]  Alexander Tropsha,et al.  Materials Informatics , 2019, J. Chem. Inf. Model..

[18]  Stéphane Mallat,et al.  Wavelet Scattering Regression of Quantum Chemical Energies , 2016, Multiscale Model. Simul..

[19]  Ekin D. Cubuk,et al.  Holistic computational structure screening of more than 12 000 candidates for solid lithium-ion conductor materials , 2017 .

[20]  Steven G. Louie,et al.  Nonlinear ionic pseudopotentials in spin-density-functional calculations , 1982 .

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

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

[23]  B. Meredig,et al.  Materials science with large-scale data and informatics: Unlocking new opportunities , 2016 .

[24]  Lorenzo Torresani,et al.  Learning Spatiotemporal Features with 3D Convolutional Networks , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[25]  J. A. Van Vechten,et al.  Quantum Dielectric Theory of Electronegativity in Covalent Systems. II. Ionization Potentials and Interband Transition Energies , 1969 .

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

[27]  Mark E. Oxley,et al.  Binary, ternary and quaternary compound former/nonformer prediction via Mendeleev number , 2001 .

[28]  Alexie M. Kolpak,et al.  Discovering charge density functionals and structure-property relationships with PROPhet: A general framework for coupling machine learning and first-principles methods , 2017, Scientific Reports.

[29]  Tianqi Chen,et al.  Empirical Evaluation of Rectified Activations in Convolutional Network , 2015, ArXiv.

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

[31]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[32]  Phillip B. Messersmith,et al.  Bioinspired antifouling polymers , 2005 .

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

[34]  R. Martin,et al.  Electronic Structure: Basic Theory and Practical Methods , 2004 .

[35]  Nongnuch Artrith,et al.  An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for TiO2 , 2016 .

[36]  Richard M. Martin.  Electronic Structure: Frontmatter , 2004 .

[37]  S. Ong,et al.  New opportunities for materials informatics: Resources and data mining techniques for uncovering hidden relationships , 2016 .

[38]  Atsuto Seko,et al.  Prediction of Low-Thermal-Conductivity Compounds with First-Principles Anharmonic Lattice-Dynamics Calculations and Bayesian Optimization. , 2015, Physical review letters.

[39]  Sebastian Scherer,et al.  VoxNet: A 3D Convolutional Neural Network for real-time object recognition , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[40]  Felix A Faber,et al.  Crystal structure representations for machine learning models of formation energies , 2015, 1503.07406.

[41]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[42]  P. Cochat,et al.  Et al , 2008, Archives de pediatrie : organe officiel de la Societe francaise de pediatrie.

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

[44]  Andrew Y. Ng,et al.  Convolutional-Recursive Deep Learning for 3D Object Classification , 2012, NIPS.

[45]  Jörg Behler,et al.  Constructing high‐dimensional neural network potentials: A tutorial review , 2015 .