Matches in SemOpenAlex for { <https://semopenalex.org/work/W4382323477> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4382323477 abstract "Recent advances in graph neural networks (GNNs) have allowed molecular simulations with accuracy on par with conventional gold-standard methods at a fraction of the computational cost. Nonetheless, as the field has been progressing to bigger and more complex architectures, state-of-the-art GNNs have become largely prohibitive for many large-scale applications. In this paper, we, for the first time, explore the utility of knowledge distillation (KD) for accelerating molecular GNNs. To this end, we devise KD strategies that facilitate the distillation of hidden representations in directional and equivariant GNNs and evaluate their performance on the regression task of energy and force prediction. We validate our protocols across different teacher-student configurations and demonstrate that they can boost the predictive accuracy of student models without altering their architecture. We also conduct comprehensive optimization of various components of our framework, and investigate the potential of data augmentation to further enhance performance. All in all, we manage to close as much as 59% of the gap in predictive accuracy between models like GemNet-OC and PaiNN with zero additional cost at inference." @default.
- W4382323477 created "2023-06-28" @default.
- W4382323477 creator A5029483098 @default.
- W4382323477 creator A5031804527 @default.
- W4382323477 creator A5082237590 @default.
- W4382323477 creator A5092346592 @default.
- W4382323477 date "2023-06-26" @default.
- W4382323477 modified "2023-09-25" @default.
- W4382323477 title "Accelerating Molecular Graph Neural Networks via Knowledge Distillation" @default.
- W4382323477 doi "https://doi.org/10.48550/arxiv.2306.14818" @default.
- W4382323477 hasPublicationYear "2023" @default.
- W4382323477 type Work @default.
- W4382323477 citedByCount "0" @default.
- W4382323477 crossrefType "posted-content" @default.
- W4382323477 hasAuthorship W4382323477A5029483098 @default.
- W4382323477 hasAuthorship W4382323477A5031804527 @default.
- W4382323477 hasAuthorship W4382323477A5082237590 @default.
- W4382323477 hasAuthorship W4382323477A5092346592 @default.
- W4382323477 hasBestOaLocation W43823234771 @default.
- W4382323477 hasConcept C119857082 @default.
- W4382323477 hasConcept C127413603 @default.
- W4382323477 hasConcept C132525143 @default.
- W4382323477 hasConcept C154945302 @default.
- W4382323477 hasConcept C178790620 @default.
- W4382323477 hasConcept C185592680 @default.
- W4382323477 hasConcept C201995342 @default.
- W4382323477 hasConcept C204030448 @default.
- W4382323477 hasConcept C2776214188 @default.
- W4382323477 hasConcept C2780451532 @default.
- W4382323477 hasConcept C41008148 @default.
- W4382323477 hasConcept C50644808 @default.
- W4382323477 hasConcept C80444323 @default.
- W4382323477 hasConceptScore W4382323477C119857082 @default.
- W4382323477 hasConceptScore W4382323477C127413603 @default.
- W4382323477 hasConceptScore W4382323477C132525143 @default.
- W4382323477 hasConceptScore W4382323477C154945302 @default.
- W4382323477 hasConceptScore W4382323477C178790620 @default.
- W4382323477 hasConceptScore W4382323477C185592680 @default.
- W4382323477 hasConceptScore W4382323477C201995342 @default.
- W4382323477 hasConceptScore W4382323477C204030448 @default.
- W4382323477 hasConceptScore W4382323477C2776214188 @default.
- W4382323477 hasConceptScore W4382323477C2780451532 @default.
- W4382323477 hasConceptScore W4382323477C41008148 @default.
- W4382323477 hasConceptScore W4382323477C50644808 @default.
- W4382323477 hasConceptScore W4382323477C80444323 @default.
- W4382323477 hasLocation W43823234771 @default.
- W4382323477 hasOpenAccess W4382323477 @default.
- W4382323477 hasPrimaryLocation W43823234771 @default.
- W4382323477 hasRelatedWork W2888706357 @default.
- W4382323477 hasRelatedWork W2961085424 @default.
- W4382323477 hasRelatedWork W2963058055 @default.
- W4382323477 hasRelatedWork W3139398652 @default.
- W4382323477 hasRelatedWork W4285260836 @default.
- W4382323477 hasRelatedWork W4286629047 @default.
- W4382323477 hasRelatedWork W4306321456 @default.
- W4382323477 hasRelatedWork W4306674287 @default.
- W4382323477 hasRelatedWork W1629725936 @default.
- W4382323477 hasRelatedWork W4224009465 @default.
- W4382323477 isParatext "false" @default.
- W4382323477 isRetracted "false" @default.
- W4382323477 workType "article" @default.