Matches in SemOpenAlex for { <https://semopenalex.org/work/W4310885979> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W4310885979 abstract "Atomistic machine learning focuses on the creation of models which obey fundamental symmetries of atomistic configurations, such as permutation, translation, and rotation invariances. In many of these schemes, translation and rotation invariance are achieved by building on scalar invariants, e.g., distances between atom pairs. There is growing interest in molecular representations that work internally with higher rank rotational tensors, e.g., vector displacements between atoms, and tensor products thereof. Here we present a framework for extending the Hierarchically Interacting Particle Neural Network (HIP-NN) with Tensor Sensitivity information (HIP-NN-TS) from each local atomic environment. Crucially, the method employs a weight tying strategy that allows direct incorporation of many-body information while adding very few model parameters. We show that HIP-NN-TS is more accurate than HIP-NN, with negligible increase in parameter count, for several datasets and network sizes. As the dataset becomes more complex, tensor sensitivities provide greater improvements to model accuracy. In particular, HIP-NN-TS achieves a record mean absolute error of 0.927 kcal/mol for conformational energy variation on the challenging COMP6 benchmark, which includes a broad set of organic molecules. We also compare the computational performance of HIP-NN-TS to HIP-NN and other models in the literature." @default.
- W4310885979 created "2022-12-20" @default.
- W4310885979 creator A5003874504 @default.
- W4310885979 creator A5004656368 @default.
- W4310885979 creator A5021344986 @default.
- W4310885979 creator A5027179168 @default.
- W4310885979 creator A5058812809 @default.
- W4310885979 creator A5059593993 @default.
- W4310885979 creator A5081624801 @default.
- W4310885979 date "2022-12-06" @default.
- W4310885979 modified "2023-10-14" @default.
- W4310885979 title "Lightweight and Effective Tensor Sensitivity for Atomistic Neural Networks" @default.
- W4310885979 doi "https://doi.org/10.48550/arxiv.2212.03195" @default.
- W4310885979 hasPublicationYear "2022" @default.
- W4310885979 type Work @default.
- W4310885979 citedByCount "0" @default.
- W4310885979 crossrefType "posted-content" @default.
- W4310885979 hasAuthorship W4310885979A5003874504 @default.
- W4310885979 hasAuthorship W4310885979A5004656368 @default.
- W4310885979 hasAuthorship W4310885979A5021344986 @default.
- W4310885979 hasAuthorship W4310885979A5027179168 @default.
- W4310885979 hasAuthorship W4310885979A5058812809 @default.
- W4310885979 hasAuthorship W4310885979A5059593993 @default.
- W4310885979 hasAuthorship W4310885979A5081624801 @default.
- W4310885979 hasBestOaLocation W43108859791 @default.
- W4310885979 hasConcept C104317684 @default.
- W4310885979 hasConcept C105580179 @default.
- W4310885979 hasConcept C11413529 @default.
- W4310885979 hasConcept C121332964 @default.
- W4310885979 hasConcept C121864883 @default.
- W4310885979 hasConcept C127413603 @default.
- W4310885979 hasConcept C13280743 @default.
- W4310885979 hasConcept C149364088 @default.
- W4310885979 hasConcept C154945302 @default.
- W4310885979 hasConcept C155281189 @default.
- W4310885979 hasConcept C185592680 @default.
- W4310885979 hasConcept C185798385 @default.
- W4310885979 hasConcept C205649164 @default.
- W4310885979 hasConcept C21200559 @default.
- W4310885979 hasConcept C24326235 @default.
- W4310885979 hasConcept C2524010 @default.
- W4310885979 hasConcept C33923547 @default.
- W4310885979 hasConcept C41008148 @default.
- W4310885979 hasConcept C50644808 @default.
- W4310885979 hasConcept C55493867 @default.
- W4310885979 hasConcept C57691317 @default.
- W4310885979 hasConcept C74050887 @default.
- W4310885979 hasConceptScore W4310885979C104317684 @default.
- W4310885979 hasConceptScore W4310885979C105580179 @default.
- W4310885979 hasConceptScore W4310885979C11413529 @default.
- W4310885979 hasConceptScore W4310885979C121332964 @default.
- W4310885979 hasConceptScore W4310885979C121864883 @default.
- W4310885979 hasConceptScore W4310885979C127413603 @default.
- W4310885979 hasConceptScore W4310885979C13280743 @default.
- W4310885979 hasConceptScore W4310885979C149364088 @default.
- W4310885979 hasConceptScore W4310885979C154945302 @default.
- W4310885979 hasConceptScore W4310885979C155281189 @default.
- W4310885979 hasConceptScore W4310885979C185592680 @default.
- W4310885979 hasConceptScore W4310885979C185798385 @default.
- W4310885979 hasConceptScore W4310885979C205649164 @default.
- W4310885979 hasConceptScore W4310885979C21200559 @default.
- W4310885979 hasConceptScore W4310885979C24326235 @default.
- W4310885979 hasConceptScore W4310885979C2524010 @default.
- W4310885979 hasConceptScore W4310885979C33923547 @default.
- W4310885979 hasConceptScore W4310885979C41008148 @default.
- W4310885979 hasConceptScore W4310885979C50644808 @default.
- W4310885979 hasConceptScore W4310885979C55493867 @default.
- W4310885979 hasConceptScore W4310885979C57691317 @default.
- W4310885979 hasConceptScore W4310885979C74050887 @default.
- W4310885979 hasLocation W43108859791 @default.
- W4310885979 hasOpenAccess W4310885979 @default.
- W4310885979 hasPrimaryLocation W43108859791 @default.
- W4310885979 hasRelatedWork W2042136086 @default.
- W4310885979 hasRelatedWork W2059569368 @default.
- W4310885979 hasRelatedWork W2067549729 @default.
- W4310885979 hasRelatedWork W2118167427 @default.
- W4310885979 hasRelatedWork W2353089562 @default.
- W4310885979 hasRelatedWork W2382412288 @default.
- W4310885979 hasRelatedWork W2401401198 @default.
- W4310885979 hasRelatedWork W2485444220 @default.
- W4310885979 hasRelatedWork W2989915422 @default.
- W4310885979 hasRelatedWork W3135923011 @default.
- W4310885979 isParatext "false" @default.
- W4310885979 isRetracted "false" @default.
- W4310885979 workType "article" @default.