Amp: A modular approach to machine learning in atomistic simulations

By Alireza Khorshidi, Andrew A. Peterson. 2016

325 citations recorded by the source corpus.

Subfields: Atomic and Molecular Physics, and Optics, Materials Chemistry

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atomic and Molecular Physics, and Optics

1. Kristian Sommer Thygesen
2. Yuri Mishin
3. James R. Kermode
4. Karsten Wedel Jacobsen
5. T. Olsen
6. G. Em Karniadakis
7. Yu-Hang Tang
8. Dongkun Zhang

In Materials Chemistry

1. Kristian Sommer Thygesen
2. Yuri Mishin
3. T. Olsen
4. James R. Kermode
5. J. Schiøtz
6. Karsten Wedel Jacobsen
7. Bjørk Hammer
8. G. Em Karniadakis

Citations in the corpus

• An Atomistic Fingerprint Algorithm for Learning Ab Initio Molecular Force Fields (2017, methodology): “Assuming that there is smoothness in the potential energy surface of the atomistic system, one possible strategy to accelerate QM-driven molecular dynamics is to use QM calculations on only a subset of the time steps, and to interpolate for similar atomic configurations [11–14].”
• Machine-Learning Interatomic Potentials for Materials Science (2021, methodology): “Amp (Atomistic Machine-learning Package) [95,140] is somehat similar to ASE and additionally contains modules for ML poential training using an assortment of descriptors and optimizaion algorithms.”
• Recent advances and applications of deep learning methods in materials science (2021, unknown): “org/andrewpeterson/amp [98] TensorMol Mol https://github.”
• Neuroevolution machine learning potentials: Combining high accuracy and low cost in atomistic simulations and application to heat transport (2021, unknown): “There are libraries implementing various descriptors [37–39].”