A5092615651- Machine Learning in Materials Science
- Electron and X-Ray Spectroscopy Techniques
- Material Science and Thermodynamics
- Surface Roughness and Optical Measurements
- Advanced Materials Characterization Techniques
- Crystallography and molecular interactions
- Advanced Electron Microscopy Techniques and Applications
- X-ray Diffraction in Crystallography
- Thermodynamic and Structural Properties of Metals and Alloys
- Glass properties and applications
- Quasicrystal Structures and Properties
- Metallic Glasses and Amorphous Alloys
- Advanced Theoretical and Applied Studies in Material Sciences and Geometry
Technical University of Darmstadt
2023-2024
Abstract We present a comprehensive and user-friendly framework built upon the integrated development environment (IDE), enabling researchers to perform entire Machine Learning Potential (MLP) cycle consisting of (i) creating systematic DFT databases, (ii) fitting Density Functional Theory (DFT) data empirical potentials or MLPs, (iii) validating in largely automatic approach. The power performance this are demonstrated for three conceptually very different classes interatomic potentials: an...
We present a comprehensive and user-friendly framework built upon the pyiron integrated development environment (IDE), enabling researchers to perform entire Machine Learning Potential (MLP) cycle consisting of (i) creating systematic DFT databases, (ii) fitting Density Functional Theory (DFT) data empirical potentials or MLPs, (iii) validating in largely automatic approach. The power performance this are demonstrated for three conceptually very different classes interatomic potentials: an...
Abstract Silicon oxycarbides show outstanding versatility due to their highly tunable composition and microstructure. Consequently, a key challenge is thorough knowledge of structure–property relations in the system. In this work, we fit an atomic cluster expansion potential set actively learned density‐functional theory training data spanning wide configurational space. We demonstrate ability produce realistic amorphous structures rationalize formation different morphologies turbostratic...
Silicon oxycarbides show outstanding versatility due to their highly tunable composition and microstructure. Consequently, a key challenge is thorough knowledge of structure-property relations in the system. In this work, we fit an atomic cluster expansion potential set actively learned DFT training data spanning wide configurational space. We demonstrate ability produce realistic amorphous structures rationalize formation different morphologies turbostratic free carbon phase. Finally,...
A general purpose machine-learning interatomic potential (MLIP) for the Cu-Zr system is presented based on atomic cluster expansion formalism [R. Drautz, ]. By using an extensive set of training data generated withdensity functional theory, this describes a wide range properties crystalline as well amorphous phases within whole compositional range. Therefore, machine learning can reproduce experimental phase diagram and structure with considerably improved accuracy. massively different...
<title>Abstract</title> We present a comprehensive and user-friendly framework built upon the pyiron integrated development environment (IDE), enabling researchers to perform entire Machine Learning Potential (MLP) cycle consisting of (i) creating systematic DFT databases, (ii) fitting Density Functional Theory (DFT) data empirical potentials or MLPs, (iii) validating in largely automatic approach. The power performance this are demonstrated for three conceptually very different classes...
A general purpose machine learning interatomic potential (MLIP) for the Cu-Zr system is presented based on Atomic Cluster Expansion (ACE) formalism (Drautz, PRB 100, 2019). By using an extensive set of training data generated with Density-Functional Theory (DFT) this describes a wide range properties crystalline as well amorphous phases within whole compositional range. Therefore, MLIP can reproduce experimental phase diagram and structure unprecedented accuracy. massively different...