Machine-learned electron correlation model based on correlation energy density at complete basis set limit.

We propose a machine-learned correlation model that is built using the regression between density variables such as electron density and correlation energy density. The correlation energy density of coupled cluster singles, doubles, and perturbative triples [CCSD(T)] is derived based on grid-based energy density analysis. The complete basis set (CBS) limit is estimated using the composite method, which has been reported to calculate the total correlation energy. The numerical examination revealed that the correlation energy density of the CCSD(T)/CBS level is appropriate for the response variable of machine learning. In addition to the density variables used in the exchange-correlation functionals of the density functional theory, the Hartree-Fock (HF) exchange energy density and electron density based on the fractional occupation number of molecular orbitals were employed as explanatory variables. Numerical assessments confirmed the accuracy and efficiency of the present correlation model. Consequently, the present protocol, namely, learning the CCSD(T)/CBS correlation energy density using density variables obtained by the HF calculation with a small basis set, yields an efficient correlation model.

[1]  M. Plesset,et al.  Note on an Approximation Treatment for Many-Electron Systems , 1934 .

[2]  P. Löwdin.  Quantum Theory of Many-Particle Systems. III. Extension of the Hartree-Fock Scheme to Include Degenerate Systems and Correlation Effects , 1955 .

[3]  R. Lefebvre,et al.  Advances in Chemical Physics: LeFebvre/Advances , 1969 .

[4]  J. Pople,et al.  Self‐consistent molecular orbital methods. XX. A basis set for correlated wave functions , 1980 .

[5]  Timothy Clark,et al.  Efficient diffuse function‐augmented basis sets for anion calculations. III. The 3‐21+G basis set for first‐row elements, Li–F , 1983 .

[6]  Michael J. Frisch,et al.  Self‐consistent molecular orbital methods 25. Supplementary functions for Gaussian basis sets , 1984 .

[7]  Werner Kutzelnigg,et al.  r12-Dependent terms in the wave function as closed sums of partial wave amplitudes for large l , 1985 .

[8]  S. J. Cole,et al.  Towards a full CCSDT model for electron correlation , 1985 .

[9]  Richard A. Friesner,et al.  Solution of self-consistent field electronic structure equations by a pseudospectral method , 1985 .

[10]  W. Kutzelnigg,et al.  Møller-plesset calculations taking care of the correlation CUSP , 1987 .

[11]  A. Becke.  A multicenter numerical integration scheme for polyatomic molecules , 1988 .

[12]  M. Head‐Gordon,et al.  A fifth-order perturbation comparison of electron correlation theories , 1989 .

[13]  T. H. Dunning.  Gaussian basis sets for use in correlated molecular calculations. I. The atoms boron through neon and hydrogen , 1989 .

[14]  Wim Klopper,et al.  Wave functions with terms linear in the interelectronic coordinates to take care of the correlation cusp. I. General theory , 1991 .

[15]  David Feller,et al.  Application of systematic sequences of wave functions to the water dimer , 1992 .

[16]  T. Dunning,et al.  Electron affinities of the first‐row atoms revisited. Systematic basis sets and wave functions , 1992 .

[17]  A. Becke.  A New Mixing of Hartree-Fock and Local Density-Functional Theories , 1993 .

[18]  Mark S. Gordon,et al.  General atomic and molecular electronic structure system , 1993, J. Comput. Chem..

[19]  C. W. Murray,et al.  Quadrature schemes for integrals of density functional theory , 1993 .

[20]  David Feller,et al.  The use of systematic sequences of wave functions for estimating the complete basis set, full configuration interaction limit in water , 1993 .

[21]  David E. Woon,et al.  Gaussian basis sets for use in correlated molecular calculations. IV. Calculation of static electrical response properties , 1994 .

[22]  Andreas Savin,et al.  Density functionals for the Yukawa electron-electron interaction , 1995 .

[23]  Thom H. Dunning,et al.  Gaussian basis sets for use in correlated molecular calculations. V. Core-valence basis sets for boron through neon , 1995 .

[24]  Michael J. Frisch,et al.  Achieving linear scaling in exchange-correlation density functional quadratures , 1996 .

[25]  Trygve Helgaker,et al.  Basis-set convergence of correlated calculations on water , 1997 .

[26]  K. Hirao,et al.  A new spin-polarized Colle-Salvetti-type correlation energy functional , 1997 .

[27]  Kimihiko Hirao,et al.  A NEW ONE-PARAMETER PROGRESSIVE COLLE-SALVETTI-TYPE CORRELATION FUNCTIONAL , 1999 .

[28]  M. Le.  A note on Jeśmanowicz' conjecture concerning Pythagorean triples , 1999, Bulletin of the Australian Mathematical Society.

[29]  David,et al.  Gaussian basis sets for use in correlated molecular calculations . Ill . The atoms aluminum through argon , 1999 .

[30]  L. Curtiss,et al.  Gaussian-3X (G3X) theory : use of improved geometries, zero-point energies, and Hartree-Fock basis sets. , 2001 .

[31]  K. Hirao,et al.  A long-range correction scheme for generalized-gradient-approximation exchange functionals , 2001 .

[32]  Karol Kowalski,et al.  Efficient computer implementation of the renormalized coupled-cluster methods: The R-CCSD[T], R-CCSD(T), CR-CCSD[T], and CR-CCSD(T) approaches , 2002 .

[33]  Jae Shin Lee,et al.  Basis set and correlation dependent extrapolation of correlation energy , 2003 .

[34]  Gustavo E. Scuseria,et al.  Local hybrid functionals , 2003 .

[35]  Lihong Hu,et al.  Combined first-principles calculation and neural-network correction approach for heat of formation , 2003 .

[36]  F. Manby,et al.  An explicitly correlated second order Møller-Plesset theory using a frozen Gaussian geminal. , 2004, The Journal of chemical physics.

[37]  Seiichiro Ten-no,et al.  Initiation of explicitly correlated Slater-type geminal theory , 2004 .

[38]  Andreas Savin,et al.  Long-range/short-range separation of the electron-electron interaction in density functional theory , 2004 .

[39]  D. Tew,et al.  New correlation factors for explicitly correlated electronic wave functions. , 2005, The Journal of chemical physics.

[40]  Frederick R. Manby,et al.  R12 methods in explicitly correlated molecular electronic structure theory , 2006 .

[41]  G. Scuseria,et al.  Assessment of a long-range corrected hybrid functional. , 2006, The Journal of chemical physics.

[42]  H. Nakai,et al.  Grid-based energy density analysis: implementation and assessment. , 2007, The Journal of chemical physics.

[43]  Xin Xu,et al.  The X1 method for accurate and efficient prediction of heats of formation. , 2007, The Journal of chemical physics.

[44]  D. Bakowies.  Extrapolation of electron correlation energies to finite and complete basis set targets. , 2007, The Journal of chemical physics.

[45]  G. Scuseria,et al.  Exact-exchange energy density in the gauge of a semilocal density functional approximation , 2007, 0710.3354.

[46]  Xin Xu,et al.  Improving the B3LYP bond energies by using the X1 method. , 2008, The Journal of chemical physics.

[47]  M. Head‐Gordon,et al.  Long-range corrected hybrid density functionals with damped atom-atom dispersion corrections. , 2008, Physical chemistry chemical physics : PCCP.

[48]  Hiromi Nakai,et al.  Energy density analysis for second‐order Møller‐Plesset perturbation theory and coupled‐cluster theory with singles and doubles: Application to C2H4CH4 complexes , 2008, J. Comput. Chem..

[49]  David Feller,et al.  A survey of factors contributing to accurate theoretical predictions of atomization energies and molecular structures. , 2008, The Journal of chemical physics.

[50]  Roman M. Balabin,et al.  Neural network approach to quantum-chemistry data: accurate prediction of density functional theory energies. , 2009, The Journal of chemical physics.

[51]  F. Neese,et al.  Efficient, approximate and parallel Hartree–Fock and hybrid DFT calculations. A ‘chain-of-spheres’ algorithm for the Hartree–Fock exchange , 2009 .

[52]  C. Corminboeuf,et al.  Reaction enthalpies using the neural-network-based X1 approach: the important choice of input descriptors. , 2009, The journal of physical chemistry. A.

[53]  A. Varandas.  Extrapolation to the complete basis set limit without counterpoise. The pair potential of helium revisited. , 2010, The journal of physical chemistry. A.

[54]  H. Eshuis,et al.  Electron correlation methods based on the random phase approximation , 2012, Theoretical Chemistry Accounts.

[55]  Matthias Scheffler,et al.  Random-phase approximation and its applications in computational chemistry and materials science , 2012, Journal of Materials Science.

[56]  Edward F. Valeev,et al.  Explicitly correlated R12/F12 methods for electronic structure. , 2012, Chemical reviews.

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

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

[59]  M. Kaupp,et al.  Importance of the correlation contribution for local hybrid functionals: range separation and self-interaction corrections. , 2012, The Journal of chemical physics.

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

[61]  M. Rupp,et al.  Machine learning of molecular electronic properties in chemical compound space , 2013, 1305.7074.

[62]  Frank Neese,et al.  Robust fitting techniques in the chain of spheres approximation to the Fock exchange: The role of the complementary space. , 2013, The Journal of chemical physics.

[63]  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.

[64]  Jörg Kussmann,et al.  Pre-selective screening for matrix elements in linear-scaling exact exchange calculations. , 2013, The Journal of chemical physics.

[65]  A. Varandas,et al.  Narrowing the error in electron correlation calculations by basis set re-hierarchization and use of the unified singlet and triplet electron-pair extrapolation scheme: application to a test set of 106 systems. , 2014, The Journal of chemical physics.

[66]  Li Li,et al.  Understanding Machine-learned Density Functionals , 2014, ArXiv.

[67]  S. Grimme,et al.  Double‐hybrid density functionals , 2014 .

[68]  Lihong Hu,et al.  Alternative approach to chemical accuracy: a neural networks-based first-principles method for heat of formation of molecules made of H, C, N, O, F, S, and Cl. , 2014, The journal of physical chemistry. A.

[69]  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.

[70]  Hiromi Nakai,et al.  Revisiting the extrapolation of correlation energies to complete basis set limit , 2015, J. Comput. Chem..

[71]  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.

[72]  Jianming Wu,et al.  Improving B3LYP heats of formation with three‐dimensional molecular descriptors , 2016, J. Comput. Chem..

[73]  N. Su,et al.  The XYG3 type of doubly hybrid density functionals , 2016 .

[74]  Kun Yao,et al.  Kinetic Energy of Hydrocarbons as a Function of Electron Density and Convolutional Neural Networks. , 2015, Journal of chemical theory and computation.

[75]  Junji Seino,et al.  Informatics‐Based Energy Fitting Scheme for Correlation Energy at Complete Basis Set Limit , 2016, J. Comput. Chem..

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

[77]  Heather J Kulik,et al.  Resolving Transition Metal Chemical Space: Feature Selection for Machine Learning and Structure-Property Relationships. , 2017, The journal of physical chemistry. A.

[78]  Li Li,et al.  Bypassing the Kohn-Sham equations with machine learning , 2016, Nature Communications.

[79]  Kipton Barros,et al.  Learning molecular energies using localized graph kernels. , 2016, The Journal of chemical physics.

[80]  Raghunathan Ramakrishnan,et al.  Genetic Optimization of Training Sets for Improved Machine Learning Models of Molecular Properties. , 2016, The journal of physical chemistry letters.

[81]  Thomas Kjærgaard,et al.  The divide–expand–consolidate coupled cluster scheme , 2017 .

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

[83]  George E. Dahl,et al.  Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT Error. , 2017, Journal of chemical theory and computation.

[84]  Lihong Hu,et al.  Improving the Performance of Long-Range-Corrected Exchange-Correlation Functional with an Embedded Neural Network. , 2017, The journal of physical chemistry. A.

[85]  M. Head‐Gordon,et al.  Thirty years of density functional theory in computational chemistry: an overview and extensive assessment of 200 density functionals , 2017 .

[86]  Karsten Reuter,et al.  Making the Coupled Cluster Correlation Energy Machine-Learnable. , 2018, The journal of physical chemistry. A.

[87]  Sergei Manzhos,et al.  Kinetic energy densities based on the fourth order gradient expansion: performance in different classes of materials and improvement via machine learning. , 2018, Physical chemistry chemical physics : PCCP.

[88]  Geoffrey J. Gordon,et al.  Constant size descriptors for accurate machine learning models of molecular properties. , 2018, The Journal of chemical physics.

[89]  Stéphane Mallat,et al.  Solid Harmonic Wavelet Scattering for Predictions of Molecule Properties , 2018, The Journal of chemical physics.

[90]  Thomas F. Miller,et al.  Transferability in Machine Learning for Electronic Structure via the Molecular Orbital Basis. , 2018, Journal of chemical theory and computation.

[91]  M Gastegger,et al.  wACSF-Weighted atom-centered symmetry functions as descriptors in machine learning potentials. , 2017, The Journal of chemical physics.

[92]  M. Kaupp,et al.  Local hybrid functionals: Theory, implementation, and performance of an emerging new tool in quantum chemistry and beyond , 2018, WIREs Computational Molecular Science.

[93]  Ryo Nagai,et al.  Neural-network Kohn-Sham exchange-correlation potential and its out-of-training transferability. , 2018, The Journal of chemical physics.

[94]  J. Kussmann,et al.  Efficient and Linear-Scaling Seminumerical Method for Local Hybrid Density Functionals. , 2018, Journal of chemical theory and computation.

[95]  Yousung Jung,et al.  A local environment descriptor for machine-learned density functional theory at the generalized gradient approximation level. , 2018, The Journal of chemical physics.

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

[97]  Jian Sun,et al.  Size-independent neural networks based first-principles method for accurate prediction of heat of formation of fuels. , 2018, The Journal of chemical physics.

[98]  Junji Seino,et al.  Semi-local machine-learned kinetic energy density functional with third-order gradients of electron density. , 2018, The Journal of chemical physics.

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

[100]  Andrew J. Medford,et al.  Design and analysis of machine learning exchange-correlation functionals via rotationally invariant convolutional descriptors , 2019, Physical Review Materials.

[101]  Thomas F. Miller,et al.  A Universal Density Matrix Functional from Molecular Orbital-Based Machine Learning: Transferability across Organic Molecules , 2019, The Journal of chemical physics.