Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387156680> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4387156680 abstract "Recent years have seen vast progress in the development of machine learned force fields (MLFFs) based on ab-initio reference calculations. Despite achieving low test errors, the suitability of MLFFs in molecular dynamics (MD) simulations is being increasingly scrutinized due to concerns about instability. Our findings suggest a potential connection between MD simulation stability and the presence of equivariant representations in MLFFs, but their computational cost can limit practical advantages they would otherwise bring. To address this, we propose a transformer architecture called SO3krates that combines sparse equivariant representations (Euclidean variables) with a self-attention mechanism that can separate invariant and equivariant information, eliminating the need for expensive tensor products. SO3krates achieves a unique combination of accuracy, stability, and speed that enables insightful analysis of quantum properties of matter on unprecedented time and system size scales. To showcase this capability, we generate stable MD trajectories for flexible peptides and supra-molecular structures with hundreds of atoms. Furthermore, we investigate the PES topology for medium-sized chainlike molecules (e.g., small peptides) by exploring thousands of minima. Remarkably, SO3krates demonstrates the ability to strike a balance between the conflicting demands of stability and the emergence of new minimum-energy conformations beyond the training data, which is crucial for realistic exploration tasks in the field of biochemistry." @default.
- W4387156680 created "2023-09-30" @default.
- W4387156680 creator A5036654011 @default.
- W4387156680 creator A5072994165 @default.
- W4387156680 creator A5075988802 @default.
- W4387156680 creator A5087792082 @default.
- W4387156680 date "2023-09-21" @default.
- W4387156680 modified "2023-09-30" @default.
- W4387156680 title "From Peptides to Nanostructures: A Euclidean Transformer for Fast and Stable Machine Learned Force Fields" @default.
- W4387156680 doi "https://doi.org/10.48550/arxiv.2309.15126" @default.
- W4387156680 hasPublicationYear "2023" @default.
- W4387156680 type Work @default.
- W4387156680 citedByCount "0" @default.
- W4387156680 crossrefType "posted-content" @default.
- W4387156680 hasAuthorship W4387156680A5036654011 @default.
- W4387156680 hasAuthorship W4387156680A5072994165 @default.
- W4387156680 hasAuthorship W4387156680A5075988802 @default.
- W4387156680 hasAuthorship W4387156680A5087792082 @default.
- W4387156680 hasBestOaLocation W43871566801 @default.
- W4387156680 hasConcept C10803110 @default.
- W4387156680 hasConcept C112972136 @default.
- W4387156680 hasConcept C114614502 @default.
- W4387156680 hasConcept C119857082 @default.
- W4387156680 hasConcept C121332964 @default.
- W4387156680 hasConcept C129782007 @default.
- W4387156680 hasConcept C134306372 @default.
- W4387156680 hasConcept C154945302 @default.
- W4387156680 hasConcept C171036898 @default.
- W4387156680 hasConcept C184720557 @default.
- W4387156680 hasConcept C186633575 @default.
- W4387156680 hasConcept C190470478 @default.
- W4387156680 hasConcept C202444582 @default.
- W4387156680 hasConcept C2524010 @default.
- W4387156680 hasConcept C33923547 @default.
- W4387156680 hasConcept C41008148 @default.
- W4387156680 hasConcept C59593255 @default.
- W4387156680 hasConcept C62520636 @default.
- W4387156680 hasConcept C80444323 @default.
- W4387156680 hasConcept C84114770 @default.
- W4387156680 hasConceptScore W4387156680C10803110 @default.
- W4387156680 hasConceptScore W4387156680C112972136 @default.
- W4387156680 hasConceptScore W4387156680C114614502 @default.
- W4387156680 hasConceptScore W4387156680C119857082 @default.
- W4387156680 hasConceptScore W4387156680C121332964 @default.
- W4387156680 hasConceptScore W4387156680C129782007 @default.
- W4387156680 hasConceptScore W4387156680C134306372 @default.
- W4387156680 hasConceptScore W4387156680C154945302 @default.
- W4387156680 hasConceptScore W4387156680C171036898 @default.
- W4387156680 hasConceptScore W4387156680C184720557 @default.
- W4387156680 hasConceptScore W4387156680C186633575 @default.
- W4387156680 hasConceptScore W4387156680C190470478 @default.
- W4387156680 hasConceptScore W4387156680C202444582 @default.
- W4387156680 hasConceptScore W4387156680C2524010 @default.
- W4387156680 hasConceptScore W4387156680C33923547 @default.
- W4387156680 hasConceptScore W4387156680C41008148 @default.
- W4387156680 hasConceptScore W4387156680C59593255 @default.
- W4387156680 hasConceptScore W4387156680C62520636 @default.
- W4387156680 hasConceptScore W4387156680C80444323 @default.
- W4387156680 hasConceptScore W4387156680C84114770 @default.
- W4387156680 hasLocation W43871566801 @default.
- W4387156680 hasOpenAccess W4387156680 @default.
- W4387156680 hasPrimaryLocation W43871566801 @default.
- W4387156680 hasRelatedWork W2028501277 @default.
- W4387156680 hasRelatedWork W2084728259 @default.
- W4387156680 hasRelatedWork W2121713324 @default.
- W4387156680 hasRelatedWork W2949066882 @default.
- W4387156680 hasRelatedWork W3006970086 @default.
- W4387156680 hasRelatedWork W3127534742 @default.
- W4387156680 hasRelatedWork W4280551281 @default.
- W4387156680 hasRelatedWork W4298380385 @default.
- W4387156680 hasRelatedWork W4353007129 @default.
- W4387156680 hasRelatedWork W4386071549 @default.
- W4387156680 isParatext "false" @default.
- W4387156680 isRetracted "false" @default.
- W4387156680 workType "article" @default.