Advanced search

Filter By:

Journal Check one or more journals to show results from those journals only.

Choose more journals

Article type Check one or more article types to show results from those article types only.
Subject Check one or more subjects to show results from those subjects only.
Date Choose a date option to show results from those dates only.

Custom date range

Clear all filters
Sort by:
Showing 1–9 of 9 results
Advanced filters: Author: Benjamin Nebgen Clear advanced filters

  • Computational modelling of chemical systems requires a balance between accuracy and computational cost. Here the authors use transfer learning to develop a general purpose neural network potential that approaches quantum-chemical accuracy for reaction thermochemistry, isomerization, and drug-like molecular torsions.

    • Justin S. Smith
    • Benjamin T. Nebgen
    • Adrian E. Roitberg
    ResearchOpen Access
    Nature Communications
    Volume: 10, P: 1-8

  • A biasing energy derived from the uncertainty of a neural network ensemble modifies the potential energy surface in molecular dynamics simulations to rapidly discover under-represented structural regions that meaningfully augment the training data set.

    • Maksim Kulichenko
    • Kipton Barros
    • Benjamin Nebgen
    ResearchOpen Access
    Nature Computational Science
    Volume: 3, P: 230-239

  • Quantum mechanical calculations of molecular ionized states are computationally quite expensive. This work reports a successful extension of a previous deep-neural networks approach towards transferable neural-network models for predicting multiple properties of open shell anions and cations.

    • Roman Zubatyuk
    • Justin S. Smith
    • Olexandr Isayev
    ResearchOpen Access
    Nature Communications
    Volume: 12, P: 1-11

  • Atomistic simulations have a broad range of applications from drug design to materials discovery. Machine learning interatomic potentials (MLIPs) have become an efficient alternative to computationally expensive ab initio simulations. Now a general reactive MLIP (called ANI-1xnr) has been developed and validated against a broad range of condensed-phase reactive systems.

    • Shuhao Zhang
    • Małgorzata Z. Makoś
    • Justin S. Smith
    ResearchOpen Access
    Nature Chemistry
    Volume: 16, P: 727-734

  • Two-dimensional electronic spectroscopy reveals the existence of intermolecular conical intersections in molecular aggregates relevant for photovoltaics.

    • Antonietta De Sio
    • Ephraim Sommer
    • Christoph Lienau
    Research
    Nature Nanotechnology
    Volume: 16, P: 63-68

  • The accuracy of a machine-learned potential is limited by the quality and diversity of the training dataset. Here the authors propose an active learning approach to automatically construct general purpose machine-learning potentials here demonstrated for the aluminum case.

    • Justin S. Smith
    • Benjamin Nebgen
    • Kipton Barros
    ResearchOpen Access
    Nature Communications
    Volume: 12, P: 1-13

  • The prediction of interatomic potentials by machine learning is a well developed method; however, interatomic potentials account for only the energies and atomic forces and neglect other essential chemical properties. This Review showcases how other properties of interest, such as atomic charges, dipole moments, long-range effects, bond orders and parameters of reduced Hamiltonians, can also be accurately predicted using machine learning models.

    • Nikita Fedik
    • Roman Zubatyuk
    • Sergei Tretiak
    Reviews
    Nature Reviews Chemistry
    Volume: 6, P: 653-672