Matches in SemOpenAlex for { <https://semopenalex.org/work/W3159454016> ?p ?o ?g. }
- W3159454016 abstract "Machine learning (ML) algorithms have undergone an explosive development impacting every aspect of computational chemistry. To obtain reliable predictions, one needs to maintain a proper balance between the black-box nature of ML frameworks and the physics of the target properties. One of the most appealing quantum-chemical properties for regression models is the electron density, and some of us recently proposed a transferable and scalable model based on the decomposition of the density onto an atom-centered basis set. The decomposition, as well as the training of the model, is at its core a minimization of some loss function, which can be arbitrarily chosen and may lead to results of different quality. Well-studied in the context of density fitting (DF), the impact of the metric on the performance of ML models has not been analyzed yet. In this work, we compare predictions obtained using the overlap and the Coulomb-repulsion metrics for both decomposition and training. As expected, the Coulomb metric used as both the DF and ML loss functions leads to the best results for the electrostatic potential and dipole moments. The origin of this difference lies in the fact that the model is not constrained to predict densities that integrate to the exact number of electrons N. Since an a posteriori correction for the number of electrons decreases the errors, we proposed a modification of the model, where N is included directly into the kernel function, which allowed lowering of the errors on the test and out-of-sample sets." @default.
- W3159454016 created "2021-05-10" @default.
- W3159454016 creator A5007563039 @default.
- W3159454016 creator A5055934562 @default.
- W3159454016 creator A5078915288 @default.
- W3159454016 date "2021-07-09" @default.
- W3159454016 modified "2023-10-02" @default.
- W3159454016 title "Impact of quantum-chemical metrics on the machine learning prediction of electron density" @default.
- W3159454016 cites W1975180250 @default.
- W3159454016 cites W1975233918 @default.
- W3159454016 cites W1985122440 @default.
- W3159454016 cites W1996565005 @default.
- W3159454016 cites W1997684194 @default.
- W3159454016 cites W1998613997 @default.
- W3159454016 cites W2023713006 @default.
- W3159454016 cites W2024113579 @default.
- W3159454016 cites W2030976617 @default.
- W3159454016 cites W2037210960 @default.
- W3159454016 cites W2037493548 @default.
- W3159454016 cites W2038305447 @default.
- W3159454016 cites W2038496679 @default.
- W3159454016 cites W2051823879 @default.
- W3159454016 cites W2053214987 @default.
- W3159454016 cites W2055230539 @default.
- W3159454016 cites W2055834715 @default.
- W3159454016 cites W2059052476 @default.
- W3159454016 cites W2065777996 @default.
- W3159454016 cites W2069006374 @default.
- W3159454016 cites W2082719796 @default.
- W3159454016 cites W2091882337 @default.
- W3159454016 cites W2112850441 @default.
- W3159454016 cites W2129946849 @default.
- W3159454016 cites W2168299072 @default.
- W3159454016 cites W2171408801 @default.
- W3159454016 cites W2755181646 @default.
- W3159454016 cites W2756519801 @default.
- W3159454016 cites W2757878424 @default.
- W3159454016 cites W2794822253 @default.
- W3159454016 cites W2889703828 @default.
- W3159454016 cites W2921533983 @default.
- W3159454016 cites W2963464916 @default.
- W3159454016 cites W2972006524 @default.
- W3159454016 cites W3017396455 @default.
- W3159454016 cites W3099739348 @default.
- W3159454016 cites W3099950071 @default.
- W3159454016 cites W3109094470 @default.
- W3159454016 doi "https://doi.org/10.1063/5.0055393" @default.
- W3159454016 hasPublicationYear "2021" @default.
- W3159454016 type Work @default.
- W3159454016 sameAs 3159454016 @default.
- W3159454016 citedByCount "1" @default.
- W3159454016 countsByYear W31594540162022 @default.
- W3159454016 crossrefType "journal-article" @default.
- W3159454016 hasAuthorship W3159454016A5007563039 @default.
- W3159454016 hasAuthorship W3159454016A5055934562 @default.
- W3159454016 hasAuthorship W3159454016A5078915288 @default.
- W3159454016 hasBestOaLocation W31594540162 @default.
- W3159454016 hasConcept C105795698 @default.
- W3159454016 hasConcept C11413529 @default.
- W3159454016 hasConcept C119857082 @default.
- W3159454016 hasConcept C121332964 @default.
- W3159454016 hasConcept C121864883 @default.
- W3159454016 hasConcept C125485243 @default.
- W3159454016 hasConcept C132459708 @default.
- W3159454016 hasConcept C147120987 @default.
- W3159454016 hasConcept C151730666 @default.
- W3159454016 hasConcept C162324750 @default.
- W3159454016 hasConcept C169903167 @default.
- W3159454016 hasConcept C173523689 @default.
- W3159454016 hasConcept C176217482 @default.
- W3159454016 hasConcept C185429906 @default.
- W3159454016 hasConcept C185592680 @default.
- W3159454016 hasConcept C21547014 @default.
- W3159454016 hasConcept C2779343474 @default.
- W3159454016 hasConcept C33923547 @default.
- W3159454016 hasConcept C41008148 @default.
- W3159454016 hasConcept C62520636 @default.
- W3159454016 hasConcept C71134354 @default.
- W3159454016 hasConcept C86803240 @default.
- W3159454016 hasConceptScore W3159454016C105795698 @default.
- W3159454016 hasConceptScore W3159454016C11413529 @default.
- W3159454016 hasConceptScore W3159454016C119857082 @default.
- W3159454016 hasConceptScore W3159454016C121332964 @default.
- W3159454016 hasConceptScore W3159454016C121864883 @default.
- W3159454016 hasConceptScore W3159454016C125485243 @default.
- W3159454016 hasConceptScore W3159454016C132459708 @default.
- W3159454016 hasConceptScore W3159454016C147120987 @default.
- W3159454016 hasConceptScore W3159454016C151730666 @default.
- W3159454016 hasConceptScore W3159454016C162324750 @default.
- W3159454016 hasConceptScore W3159454016C169903167 @default.
- W3159454016 hasConceptScore W3159454016C173523689 @default.
- W3159454016 hasConceptScore W3159454016C176217482 @default.
- W3159454016 hasConceptScore W3159454016C185429906 @default.
- W3159454016 hasConceptScore W3159454016C185592680 @default.
- W3159454016 hasConceptScore W3159454016C21547014 @default.
- W3159454016 hasConceptScore W3159454016C2779343474 @default.
- W3159454016 hasConceptScore W3159454016C33923547 @default.
- W3159454016 hasConceptScore W3159454016C41008148 @default.
- W3159454016 hasConceptScore W3159454016C62520636 @default.
- W3159454016 hasConceptScore W3159454016C71134354 @default.