Size‐Extensive Molecular Machine Learning with Global Representations †
Abstract: Machine learning (ML) models are increasingly used in combination with electronic structure calculations to predict molecular properties at a much lower computational cost in high‐throughput settings. Such ML models require representations that encode the molecular structure, which are generally designed to respect the symmetries and invariances of the target property. However, size‐extensivity is usually not guaranteed for so‐called global representations. In this contribution, we show how extensivity can be built into global ML models using, e. g., the Many‐Body Tensor Representation. Properties of extensive and non‐extensive models for the atomization energy are systematically explored by training on small molecules and testing on small, medium and large molecules. Our results show that non‐extensive models are only useful in the size‐range of their training set, whereas extensive models provide reasonable predictions across large size differences. Remaining sources of error for extensive models are discussed.
- Location
-
Deutsche Nationalbibliothek Frankfurt am Main
- Extent
-
Online-Ressource
- Language
-
Englisch
- Bibliographic citation
-
Size‐Extensive Molecular Machine Learning with Global Representations † ; volume:2 ; number:4 ; year:2020 ; extent:7
ChemSystemsChem ; 2, Heft 4 (2020) (gesamt 7)
- Creator
-
Jung, Hyunwook
Stocker, Sina
Kunkel, Christian
Oberhofer, Harald
Han, Byungchan
Reuter, Karsten
Margraf, Johannes T.
- DOI
-
10.1002/syst.201900052
- URN
-
urn:nbn:de:101:1-2022062910340301146108
- Rights
-
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
- Last update
-
15.08.2025, 7:31 AM CEST
Data provider
This object is provided by:
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Associated
- Jung, Hyunwook
- Stocker, Sina
- Kunkel, Christian
- Oberhofer, Harald
- Han, Byungchan
- Reuter, Karsten
- Margraf, Johannes T.
Other Objects (12)
Pure non-local machine-learned density functional theory for electron correlation
First-Principles Approach to Heat and Mass Transfer Effects in Model Catalyst Studies
Epitaxial Core‐Shell Oxide Nanoparticles: First‐Principles Evidence for Increased Activity and Stability of Rutile Catalysts for Acidic Oxygen Evolution
Discoveries in Ruthenium Oxide-Based Catalysts: From Morphology Control for Water Electrolysis to Surface Structure Determination via Machine-Learning
Machine learning in chemical reaction space
Multiscale Modeling of Disorder in Solid-State Battery Materials
Shape and Catalytic Mechanism of RuO2 Particles at CO Oxidation Reaction Conditions: First-Principles Based Multi-Scale Modeling
First-principles Multi-scale Simulations of Dynamic Catalyst Surfaces : CO Oxidation from Palladium Surface Oxide to Palladium Metal
Response properties at dynamic liquid-liquid interfaces
In Silico Prediction of Dissolution Rates of Pharmaceutical Ingredients: Micro-Kinetic Model Based on Spiral Dissolution
Free-base and metalated porphyrins on metal surfaces - a systematic X-ray spectroscopy and density functional theory investigation
First-Principles Microkinetic and Computational Screening Study on Doping as a Means to Improve the Deacon Process at RuO2
Pure non-local machine-learned density functional theory for electron correlation
First-Principles Approach to Heat and Mass Transfer Effects in Model Catalyst Studies
Epitaxial Core‐Shell Oxide Nanoparticles: First‐Principles Evidence for Increased Activity and Stability of Rutile Catalysts for Acidic Oxygen Evolution
Discoveries in Ruthenium Oxide-Based Catalysts: From Morphology Control for Water Electrolysis to Surface Structure Determination via Machine-Learning
Machine learning in chemical reaction space
Multiscale Modeling of Disorder in Solid-State Battery Materials
Shape and Catalytic Mechanism of RuO2 Particles at CO Oxidation Reaction Conditions: First-Principles Based Multi-Scale Modeling
First-principles Multi-scale Simulations of Dynamic Catalyst Surfaces : CO Oxidation from Palladium Surface Oxide to Palladium Metal
Response properties at dynamic liquid-liquid interfaces
In Silico Prediction of Dissolution Rates of Pharmaceutical Ingredients: Micro-Kinetic Model Based on Spiral Dissolution
Free-base and metalated porphyrins on metal surfaces - a systematic X-ray spectroscopy and density functional theory investigation
First-Principles Microkinetic and Computational Screening Study on Doping as a Means to Improve the Deacon Process at RuO2
Pure non-local machine-learned density functional theory for electron correlation
First-Principles Approach to Heat and Mass Transfer Effects in Model Catalyst Studies
Epitaxial Core‐Shell Oxide Nanoparticles: First‐Principles Evidence for Increased Activity and Stability of Rutile Catalysts for Acidic Oxygen Evolution
Discoveries in Ruthenium Oxide-Based Catalysts: From Morphology Control for Water Electrolysis to Surface Structure Determination via Machine-Learning
Machine learning in chemical reaction space
Multiscale Modeling of Disorder in Solid-State Battery Materials
Shape and Catalytic Mechanism of RuO2 Particles at CO Oxidation Reaction Conditions: First-Principles Based Multi-Scale Modeling
First-principles Multi-scale Simulations of Dynamic Catalyst Surfaces : CO Oxidation from Palladium Surface Oxide to Palladium Metal
Response properties at dynamic liquid-liquid interfaces
In Silico Prediction of Dissolution Rates of Pharmaceutical Ingredients: Micro-Kinetic Model Based on Spiral Dissolution
Free-base and metalated porphyrins on metal surfaces - a systematic X-ray spectroscopy and density functional theory investigation