Unsupervised learning of atomic environments from simple features

@article{Reinhart2021UnsupervisedLO,
  title={Unsupervised learning of atomic environments from simple features},
  author={Wesley F. Reinhart},
  journal={Computational Materials Science},
  year={2021},
  volume={196},
  pages={110511}
}
  • Wesley F. Reinhart
  • Published 28 February 2021
  • Computer Science
  • Computational Materials Science
5 Citations
Predicting aggregation of sequence-defined macromolecules with Recurrent Neural Networks
TLDR
It is shown that regression models with implicit representation learning perform significantly better than those based on engineered features such as k -mer counting, and a Recurrent-Neural-Network-based regressor performs the best.
Predicting aggregate morphology of sequence-defined macromolecules with recurrent neural networks.
TLDR
This work accurately predicts the morphology of aggregates self-assembled from sequence-defined macromolecules using supervised machine learning and demonstrates the high-throughput screening of monomer sequences using the regression model to identify candidates for self-assembly into selected morphologies.
Unsupervised learning of sequence-specific aggregation behavior for a model copolymer.
TLDR
This work applies a recently developed unsupervised machine learning scheme for local environments to characterize large-scale, disordered aggregates formed by sequence-defined macromolecules, and finds that the spatiotemporal evolution of systems in the learned latent space is smooth and continuous.

References

SHOWING 1-10 OF 40 REFERENCES
A universal strategy for the creation of machine learning-based atomistic force fields
TLDR
A general and universal strategy for using machine learning-based methods to generate highly accurate atomic force fields that may provide a pathway for performing efficient molecular dynamics simulations on nanometer-sized systems over several nanoseconds.
Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties.
TLDR
A crystal graph convolutional neural networks framework to directly learn material properties from the connection of atoms in the crystal, providing a universal and interpretable representation of crystalline materials.
On representing chemical environments
TLDR
It is demonstrated that certain widely used descriptors that initially look quite different are specific cases of a general approach, in which a finite set of basis functions with increasing angular wave numbers are used to expand the atomic neighborhood density function.
Machine learning for autonomous crystal structure identification.
TLDR
This work uses nonlinear manifold learning to infer structural relationships between particles according to the topology of their local environment, which yields unbiased structural information which allows them to quantify the crystalline character of particles near defects, grain boundaries, and interfaces.
Efficient and accurate machine-learning interpolation of atomic energies in compositions with many species
TLDR
It is shown that a mathematically simple and computationally efficient descriptor with constant complexity is sufficient to represent transition-metal oxide compositions and biomolecules containing 11 chemical species with a precision of around 3 meV/atom.
Robust structural identification via polyhedral template matching
Successful scientific applications of large-scale molecular dynamics often rely on automated methods for identifying the local crystalline structure of condensed phases. Many existing methods for
Multi-atom pattern analysis for binary superlattices.
TLDR
This work develops a framework for the analysis of multi-atom patterns, which incorporate structural information from the second coordination shell while providing a unified signature for all constituent particles in the superlattice, and constructs an efficient metric for making quantitative comparisons between these patterns.
Solving the electronic structure problem with machine learning
TLDR
A machine-learning-based scheme to efficiently assimilate the function of the KS equation, and by-pass it to directly, rapidly, and accurately predict the electronic structure of a material or a molecule, given just its atomic configuration is introduced.
Learning scheme to predict atomic forces and accelerate materials simulations
TLDR
It is argued that pathways are available to systematically and continuously improve the predictive capability of such a learned force field in an adaptive manner, and that this concept can be generalized to include multiple elements.
End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems
TLDR
Deep Potential - Smooth Edition (DeepPot-SE), an end-to-end machine learning-based PES model, which is able to efficiently represent the PES for a wide variety of systems with the accuracy of ab initio quantum mechanics models is developed.
...
...