Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Hierarchical modeling of molecular energies using a deep neural network

Journal Article · · Journal of Chemical Physics
DOI:https://doi.org/10.1063/1.5011181· OSTI ID:1479929
 [1];  [2];  [1]
  1. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
  2. Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Univ. of Florida, Gainesville, FL (United States)
+ Show Author Affiliations
In this work, we introduce the Hierarchically Interacting Particle Neural Network (HIP-NN) to model molecular properties from datasets of quantum calculations. Inspired by a many-body expansion, HIP-NN decomposes properties, such as energy, as a sum over hierarchical terms. These terms are generated from a neural network—a composition of many nonlinear transformations—acting on a representation of the molecule. HIP-NN achieves the state-of-the-art performance on a dataset of 131k ground state organic molecules and predicts energies with 0.26 kcal/mol mean absolute error. Finally, with minimal tuning, our model is also competitive on a dataset of molecular dynamics trajectories. In addition to enabling accurate energy predictions, the hierarchical structure of HIP-NN helps to identify regions of model uncertainty.
Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Sponsoring Organization:
Laboratory Directed Research and Development (LDRD); USDOE
Grant/Contract Number:
AC52-06NA25396
OSTI ID:
1479929
Alternate ID(s):
OSTI ID: 1426848
Report Number(s):
LA-UR--17-28890
Journal Information:
Journal of Chemical Physics, Journal Name: Journal of Chemical Physics Journal Issue: 24 Vol. 148; ISSN 0021-9606
Publisher:
American Institute of Physics (AIP)Copyright Statement
Country of Publication:
United States
Language:
English

References (59)

IDH mutations may not preclude distant, trans-tentorial spread in gliomas: a case report and review of the literature journal February 2016
On Combining Different Acceleration Techniques at the Iterative Solution of PDEs by the Method of Collocations and Least Residuals journal January 2017
Constructing high-dimensional neural network potentials: A tutorial review journal March 2015
Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties
  • von Lilienfeld, O. Anatole; Ramakrishnan, Raghunathan; Rupp, Matthias
  • International Journal of Quantum Chemistry, Vol. 115, Issue 16 https://doi.org/10.1002/qua.24912
journal April 2015
Gaussian approximation potentials: A brief tutorial introduction journal April 2015
Density Functional Theory book January 2011
Density Functional Theory book January 1990
Molecular graph convolutions: moving beyond fingerprints journal August 2016
Active learning of linearly parametrized interatomic potentials journal December 2017
Efficient Parallel Linear Scaling Construction of the Density Matrix for Born–Oppenheimer Molecular Dynamics journal September 2015
Structural Ensembles of Intrinsically Disordered Proteins Depend Strongly on Force Field: A Comparison to Experiment journal October 2015
Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT Error journal October 2017
Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space journal June 2015
Machine Learning for Quantum Mechanical Properties of Atoms in Molecules journal July 2015
Intrinsic Bond Energies from a Bonds-in-Molecules Neural Network journal June 2017
Enumeration of 166 Billion Organic Small Molecules in the Chemical Universe Database GDB-17 journal November 2012
Many-Body Effects in Intermolecular Forces journal November 1994
ReaxFF:  A Reactive Force Field for Hydrocarbons journal October 2001
Deep learning journal May 2015
Quantum-chemical insights from deep tensor neural networks journal January 2017
The ReaxFF reactive force-field: development, applications and future directions journal March 2016
Quantum chemistry structures and properties of 134 kilo molecules journal August 2014
Comparing molecules and solids across structural and alchemical space journal January 2016
ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost journal January 2017
On the Correlation Problem in Atomic and Molecular Systems. Calculation of Wavefunction Components in Ursell‐Type Expansion Using Quantum‐Field Theoretical Methods journal December 1966
Machine learning scheme for fast extraction of chemically interpretable interatomic potentials journal August 2016
The many-body expansion combined with neural networks journal January 2017
Learning molecular energies using localized graph kernels journal March 2017
Density functional tight binding
  • Elstner, Marcus; Seifert, Gotthard
  • Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, Vol. 372, Issue 2011 https://doi.org/10.1098/rsta.2012.0483
journal March 2014
Self-Consistent Equations Including Exchange and Correlation Effects journal November 1965
Computer simulation of local order in condensed phases of silicon journal April 1985
Self-consistent-charge density-functional tight-binding method for simulations of complex materials properties journal September 1998
Publisher’s Note: On representing chemical environments [Phys. Rev. B 87 , 184115 (2013)] journal June 2013
Machine learning based interatomic potential for amorphous carbon journal March 2017
Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species journal July 2017
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning journal January 2012
Molecular Dynamics with On-the-Fly Machine Learning of Quantum-Mechanical Forces journal March 2015
Generalized Gradient Approximation Made Simple journal October 1996
Generalized Neural-Network Representation of High-Dimensional Potential-Energy Surfaces journal April 2007
Coupled-cluster theory in quantum chemistry journal February 2007
Machine learning of accurate energy-conserving molecular force fields journal May 2017
Moment Tensor Potentials: A Class of Systematically Improvable Interatomic Potentials journal January 2016
Machine learning of accurate energy-conserving molecular force fields text January 2017
ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost text January 2017
Machine learning based interatomic potential for amorphous carbon text January 2017
Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning text January 2011
Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties preprint January 2013
Gaussian Approximation Potentials: a brief tutorial introduction preprint January 2015
Comparing molecules and solids across structural and alchemical space text January 2016
Quantum-Chemical Insights from Deep Tensor Neural Networks text January 2016
Machine Learning of Accurate Energy-Conserving Molecular Force Fields text January 2016
Learning molecular energies using localized graph kernels text January 2016
Deep Learning text January 2018
Machine Learning Predictions of Molecular Properties: Accurate Many-Body Potentials and Nonlocality in Chemical Space text January 2015
Quantum chemistry structures and properties of 134 kilo molecules text January 2014
Fast and accurate modeling of molecular atomization energies with machine learning text January 2012
Machine Learning for Quantum Mechanical Properties of Atoms in Molecules text January 2015
Fourier series of atomic radial distribution functions: A molecular fingerprint for machine learning models of quantum chemical properties text January 2015
Prediction Errors of Molecular Machine Learning Models Lower than Hybrid DFT Error text January 2017

Cited By (26)

Fast and Accurate Artificial Neural Network Potential Model for MAPbI3 Perovskite Materials journal June 2019
ANI-1: A data set of 20M off-equilibrium DFT calculations for organic molecules text January 2017
Enhanced Deep‐Learning Prediction of Molecular Properties via Augmentation of Bond Topology journal August 2019
Band Gap Prediction for Large Organic Crystal Structures with Machine Learning journal July 2019
Data-enabled structure–property mappings for lanthanide-activated inorganic scintillators journal February 2019
Towards exact molecular dynamics simulations with machine-learned force fields journal September 2018
Machine-learned multi-system surrogate models for materials prediction journal April 2019
Feature optimization for atomistic machine learning yields a data-driven construction of the periodic table of the elements journal January 2018
Enumeration of de novo inorganic complexes for chemical discovery and machine learning journal January 2020
A quantitative uncertainty metric controls error in neural network-driven chemical discovery journal January 2019
Machine learning of molecular properties: Locality and active learning journal June 2018
Guest Editorial: Special Topic on Data-Enabled Theoretical Chemistry journal June 2018
Unsupervised machine learning for detection of phase transitions in off-lattice systems. II. Applications journal November 2018
Molecular force fields with gradient-domain machine learning: Construction and application to dynamics of small molecules with coupled cluster forces journal March 2019
Chemical diversity in molecular orbital energy predictions with kernel ridge regression journal May 2019
A universal density matrix functional from molecular orbital-based machine learning: Transferability across organic molecules journal April 2019
Machine-learned electron correlation model based on correlation energy density at complete basis set limit journal July 2019
Data-driven material models for atomistic simulation journal May 2019
Machine Learning a General-Purpose Interatomic Potential for Silicon journal December 2018
Machine learning and the physical sciences journal December 2019
Dataset’s chemical diversity limits the generalizability of machine learning predictions journal November 2019
Machine Learning a General-Purpose Interatomic Potential for Silicon text January 2018
Machine-learned multi-system surrogate models for materials prediction text January 2019
Machine-learned multi-system surrogate models for materials prediction text January 2018
Molecular Force Fields with Gradient-Domain Machine Learning: Construction and Application to Dynamics of Small Molecules with Coupled Cluster Forces text January 2019
Data-driven Material Models for Atomistic Simulation text January 2019

Similar Records

Lightweight and effective tensor sensitivity for atomistic neural networks
Journal Article · Mon May 08 20:00:00 EDT 2023 · Journal of Chemical Physics · OSTI ID:2441369
Bond order predictions using deep neural networks
Journal Article · Sat Feb 13 19:00:00 EST 2021 · Journal of Applied Physics · OSTI ID:1805727

Related Subjects

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY
97 MATHEMATICS AND COMPUTING