mitwood/mlearn

Benchmark Suite for Machine Learning Interatomic Potentials for Materials

mlearn

The mlearn package is a benchmark suite for machine learning interatomic potentials for materials science. It enables a seamless way to develop various potentials and provides LAMMPS-driven properties predictor with developed potentials as plugins.

Installation

The usage of mlearn requires installation of specific packages and the plugins in LAMMPS. Please see detailed installation instructions for all descriptors.

Jupyter Notebook Examples

• Gaussian Approximation Potential (GAP)
• Moment Tensor Potential (MTP)
• Neural Network Potential (NNP)
• Linear Spectral Neighbor Analysis Potential (SNAP) and quadratic SNAP

References

• Gaussian Approximation Potential: Bartók, A. P.; Payne, M. C.; Kondor, R.; Csányi, G. Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons. Physical Review Letters 2010, 104, 136403. doi:10.1103/PhysRevLett.104.136403.
• Moment Tensor Potential: Shapeev, A. V. Moment tensor potentials: A class of systematically improvable interatomic potentials. Multiscale Modeling & Simulation, 14(3), 1153-1173. doi:10.1137/15M1054183
• Neural Network Potential: Behler, J., & Parrinello, M. Generalized neural-network representation of high-dimensional potential-energy surfaces. Physical Review Letters 2007, 98, 146401. doi:10.1103/PhysRevLett.98.146401
• Spectral Neighbor Analysis Potential: Thompson, A. P., Swiler, L. P., Trott, C. R., Foiles, S. M., & Tucker, G. J. Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials. Journal of Computational Physics, 285, 316-330. doi:10.1016/j.jcp.12.018