Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386691567> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W4386691567 endingPage "502" @default.
- W4386691567 startingPage "492" @default.
- W4386691567 abstract "Machine learning is a data-driven domain, which means a learning model's performance depends on the availability of large volumes of data to train it. However, by improving data quality, we can train effective machine learning models with little data. This paper demonstrates this possibility by proposing a methodology to generate high-quality data in the networking domain. We designed a dataset to train a given Graph Neural Network (GNN) that not only contains a small number of samples, but whose samples also feature network graphs of a reduced size (10-node networks). Our evaluations indicate that the dataset generated by the proposed pipeline can train a GNN model that scales well to larger networks of 50 to 300 nodes. The trained model compares favorably to the baseline, achieving a mean absolute percentage error of 5-6%, while being significantly smaller at 90 samples total (vs. thousands of samples for the baseline)." @default.
- W4386691567 created "2023-09-13" @default.
- W4386691567 creator A5020805931 @default.
- W4386691567 creator A5037051054 @default.
- W4386691567 creator A5045673027 @default.
- W4386691567 creator A5048080334 @default.
- W4386691567 date "2023-09-12" @default.
- W4386691567 modified "2023-10-14" @default.
- W4386691567 title "Designing graph neural networks training data with limited samples and small network sizes" @default.
- W4386691567 doi "https://doi.org/10.52953/afyw5455" @default.
- W4386691567 hasPublicationYear "2023" @default.
- W4386691567 type Work @default.
- W4386691567 citedByCount "0" @default.
- W4386691567 crossrefType "journal-article" @default.
- W4386691567 hasAuthorship W4386691567A5020805931 @default.
- W4386691567 hasAuthorship W4386691567A5037051054 @default.
- W4386691567 hasAuthorship W4386691567A5045673027 @default.
- W4386691567 hasAuthorship W4386691567A5048080334 @default.
- W4386691567 hasConcept C111368507 @default.
- W4386691567 hasConcept C119857082 @default.
- W4386691567 hasConcept C124101348 @default.
- W4386691567 hasConcept C12725497 @default.
- W4386691567 hasConcept C127313418 @default.
- W4386691567 hasConcept C127413603 @default.
- W4386691567 hasConcept C132525143 @default.
- W4386691567 hasConcept C134306372 @default.
- W4386691567 hasConcept C138885662 @default.
- W4386691567 hasConcept C154945302 @default.
- W4386691567 hasConcept C199360897 @default.
- W4386691567 hasConcept C2776401178 @default.
- W4386691567 hasConcept C2779280203 @default.
- W4386691567 hasConcept C33923547 @default.
- W4386691567 hasConcept C36503486 @default.
- W4386691567 hasConcept C41008148 @default.
- W4386691567 hasConcept C41895202 @default.
- W4386691567 hasConcept C43521106 @default.
- W4386691567 hasConcept C50644808 @default.
- W4386691567 hasConcept C51632099 @default.
- W4386691567 hasConcept C62611344 @default.
- W4386691567 hasConcept C66938386 @default.
- W4386691567 hasConcept C80444323 @default.
- W4386691567 hasConceptScore W4386691567C111368507 @default.
- W4386691567 hasConceptScore W4386691567C119857082 @default.
- W4386691567 hasConceptScore W4386691567C124101348 @default.
- W4386691567 hasConceptScore W4386691567C12725497 @default.
- W4386691567 hasConceptScore W4386691567C127313418 @default.
- W4386691567 hasConceptScore W4386691567C127413603 @default.
- W4386691567 hasConceptScore W4386691567C132525143 @default.
- W4386691567 hasConceptScore W4386691567C134306372 @default.
- W4386691567 hasConceptScore W4386691567C138885662 @default.
- W4386691567 hasConceptScore W4386691567C154945302 @default.
- W4386691567 hasConceptScore W4386691567C199360897 @default.
- W4386691567 hasConceptScore W4386691567C2776401178 @default.
- W4386691567 hasConceptScore W4386691567C2779280203 @default.
- W4386691567 hasConceptScore W4386691567C33923547 @default.
- W4386691567 hasConceptScore W4386691567C36503486 @default.
- W4386691567 hasConceptScore W4386691567C41008148 @default.
- W4386691567 hasConceptScore W4386691567C41895202 @default.
- W4386691567 hasConceptScore W4386691567C43521106 @default.
- W4386691567 hasConceptScore W4386691567C50644808 @default.
- W4386691567 hasConceptScore W4386691567C51632099 @default.
- W4386691567 hasConceptScore W4386691567C62611344 @default.
- W4386691567 hasConceptScore W4386691567C66938386 @default.
- W4386691567 hasConceptScore W4386691567C80444323 @default.
- W4386691567 hasIssue "3" @default.
- W4386691567 hasLocation W43866915671 @default.
- W4386691567 hasOpenAccess W4386691567 @default.
- W4386691567 hasPrimaryLocation W43866915671 @default.
- W4386691567 hasRelatedWork W2961085424 @default.
- W4386691567 hasRelatedWork W2992516105 @default.
- W4386691567 hasRelatedWork W3046775127 @default.
- W4386691567 hasRelatedWork W3170094116 @default.
- W4386691567 hasRelatedWork W4205958290 @default.
- W4386691567 hasRelatedWork W4285260836 @default.
- W4386691567 hasRelatedWork W4286629047 @default.
- W4386691567 hasRelatedWork W4306321456 @default.
- W4386691567 hasRelatedWork W4306674287 @default.
- W4386691567 hasRelatedWork W4224009465 @default.
- W4386691567 hasVolume "4" @default.
- W4386691567 isParatext "false" @default.
- W4386691567 isRetracted "false" @default.
- W4386691567 workType "article" @default.