On-the-Fly Active Learning of Interatomic Potentials for Large-Scale Atomistic Simulations

Jinnouchi, Ryosuke; Miwa, Kenji; Karsai, Ferenc; Kresse, Georg; Asahi, Ryoji

doi:10.1021/acs.jpclett.0c01061

Cited by 124 publications

(92 citation statements)

References 94 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For a comprehensive description of the on-the-fly MLFF generation during MD simulations implemented in VASP and employed in the present work, we refer to Jinnouchi et al (2019a) and Jinnouchi et al (2019b). AL is amply discussed in, e.g., Artrith and Behler (2012), Miwa and Ohno (2017), Jacobsen et al (2018), Zhang et al (2019), and (Jinnouchi et al, 2020c).…”

Section: Models and Methodsmentioning

confidence: 99%

Machine Learning in Computational Surface Science and Catalysis: Case Studies on Water and Metal–Oxide Interfaces

Paier

2020

Front. Chem.

View full text Add to dashboard Cite

The goal of many computational physicists and chemists is the ability to bridge the gap between atomistic length scales of about a few multiples of an Ångström (Å), i. e., 10−10 m, and meso- or macroscopic length scales by virtue of simulations. The same applies to timescales. Machine learning techniques appear to bring this goal into reach. This work applies the recently published on-the-fly machine-learned force field techniques using a variant of the Gaussian approximation potentials combined with Bayesian regression and molecular dynamics as efficiently implemented in the Vienna ab initio simulation package, VASP. The generation of these force fields follows active-learning schemes. We apply these force fields to simple oxides such as MgO and more complex reducible oxides such as iron oxide, examine their generalizability, and further increase complexity by studying water adsorption on these metal oxide surfaces. We successfully examined surface properties of pristine and reconstructed MgO and Fe3O4 surfaces. However, the accurate description of water–oxide interfaces by machine-learned force fields, especially for iron oxides, remains a field offering plenty of research opportunities.

show abstract

Section: Models and Methodsmentioning

confidence: 99%

Machine Learning in Computational Surface Science and Catalysis: Case Studies on Water and Metal–Oxide Interfaces

Paier

2020

Front. Chem.

View full text Add to dashboard Cite

show abstract

“…To perform and supplement the aforementioned studies with methods and data sets, numerous software packages have been developed over recent years. We briefly mention the available codes and categorize them into three main types, the first of which being those related to the acceleration of legacy quantum codes, such as ab initio molecular dynamics (MD) runs in VASP, 392 Gaussian process based geometry optimization in ASE, 393 machine learning adaptive basis sets within CP2K, 238 as well as SNAP 394 in LAMMPS, a machine-learning interatomic potential using bispectrum components to characterize the local neighborhood of each atom of the system.…”

Section: Software Packagesmentioning

confidence: 99%

Ab Initio Machine Learning in Chemical Compound Space

2021

View full text Add to dashboard Cite

Chemical compound space (CCS), the set of all theoretically conceivable combinations of chemical elements and (meta-)stable geometries that make up matter, is colossal. The first-principles based virtual sampling of this space, for example, in search of novel molecules or materials which exhibit desirable properties, is therefore prohibitive for all but the smallest subsets and simplest properties. We review studies aimed at tackling this challenge using modern machine learning techniques based on (i) synthetic data, typically generated using quantum mechanics based methods, and (ii) model architectures inspired by quantum mechanics. Such Quantum mechanics based Machine Learning (QML) approaches combine the numerical efficiency of statistical surrogate models with an ab initio view on matter. They rigorously reflect the underlying physics in order to reach universality and transferability across CCS. While state-of-the-art approximations to quantum problems impose severe computational bottlenecks, recent QML based developments indicate the possibility of substantial acceleration without sacrificing the predictive power of quantum mechanics.

show abstract

“…46 AL schemes have also been combined with GP based force fields including GAP, 47 and included within a first-principles MD implementation such that it allows the "on the fly" fitting of force fields for a specific simulation system. 48,49 Efficient approaches to generate reactive ML potentials become even more important when exploring chemical reactions in molecular systems, which often require a description at a computational level beyond DFT, and therefore require reference data at the same level. Very recently, AL approaches have started to be adopted for fitting reactive potentials for organic molecules based on single point evaluations at quantum-chemical levels of theory.…”

Section: Introductionmentioning

confidence: 99%