Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385482734> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4385482734 abstract "Molecular classification benefits a lot from the re-cent success of graph contrastive learning (GCL) which pulls positive samples close and pushes the negative samples apart. GCL methods generate negative and positive samples via graph augmentation. Due to the structural corruption caused by graph augmentation, not all generated negative samples retain discrim-inative semantics. However, existing GCL methods ignore the difference between negative samples and hold an assumption that the importance of all negative samples is the same, leading to degraded performance of molecular classification. To address this issue, in this paper, we propose a novel molecular graph contrastive learning model (MocGCL) by selecting more useful negative samples to improve the performance of molecular classification. Specifically, we first employ different encoders to generate positive samples to improve the diversity of positive samples. Then, we design negative generation to generate negative samples and define semantic integrity to measure the usefulness of generated negative samples. Moreover, we propose the novel negative selection to dynamically select the negative samples of more usefulness to improve the molecular representation. In addition, we improve the contrastive loss to adaptively adjust the distance between selected negative samples, which can pre-serve the distinctive properties of selected negative samples in sample space. Extensive experiments on six typical bioinformatics datasets demonstrate the effectiveness of our MocGCL compared to most state-of-the-art methods." @default.
- W4385482734 created "2023-08-03" @default.
- W4385482734 creator A5007080915 @default.
- W4385482734 creator A5029400872 @default.
- W4385482734 creator A5045973629 @default.
- W4385482734 creator A5061699251 @default.
- W4385482734 creator A5069241620 @default.
- W4385482734 creator A5079053446 @default.
- W4385482734 creator A5081132555 @default.
- W4385482734 date "2023-06-18" @default.
- W4385482734 modified "2023-10-01" @default.
- W4385482734 title "MocGCL: Molecular Graph Contrastive Learning via Negative Selection" @default.
- W4385482734 cites W1874566015 @default.
- W4385482734 cites W1988037271 @default.
- W4385482734 cites W2008857988 @default.
- W4385482734 cites W2092750499 @default.
- W4385482734 cites W2109363337 @default.
- W4385482734 cites W2157511453 @default.
- W4385482734 cites W2943472942 @default.
- W4385482734 cites W2963066159 @default.
- W4385482734 cites W3035524453 @default.
- W4385482734 cites W3086452730 @default.
- W4385482734 cites W3094497296 @default.
- W4385482734 cites W3145004415 @default.
- W4385482734 cites W3153673236 @default.
- W4385482734 cites W3205655893 @default.
- W4385482734 cites W4206480231 @default.
- W4385482734 cites W4221023051 @default.
- W4385482734 cites W4225287123 @default.
- W4385482734 cites W4225310585 @default.
- W4385482734 doi "https://doi.org/10.1109/ijcnn54540.2023.10191518" @default.
- W4385482734 hasPublicationYear "2023" @default.
- W4385482734 type Work @default.
- W4385482734 citedByCount "0" @default.
- W4385482734 crossrefType "proceedings-article" @default.
- W4385482734 hasAuthorship W4385482734A5007080915 @default.
- W4385482734 hasAuthorship W4385482734A5029400872 @default.
- W4385482734 hasAuthorship W4385482734A5045973629 @default.
- W4385482734 hasAuthorship W4385482734A5061699251 @default.
- W4385482734 hasAuthorship W4385482734A5069241620 @default.
- W4385482734 hasAuthorship W4385482734A5079053446 @default.
- W4385482734 hasAuthorship W4385482734A5081132555 @default.
- W4385482734 hasConcept C104317684 @default.
- W4385482734 hasConcept C119857082 @default.
- W4385482734 hasConcept C132525143 @default.
- W4385482734 hasConcept C141231307 @default.
- W4385482734 hasConcept C153180895 @default.
- W4385482734 hasConcept C154945302 @default.
- W4385482734 hasConcept C204321447 @default.
- W4385482734 hasConcept C41008148 @default.
- W4385482734 hasConcept C55493867 @default.
- W4385482734 hasConcept C7386963 @default.
- W4385482734 hasConcept C80444323 @default.
- W4385482734 hasConcept C86803240 @default.
- W4385482734 hasConceptScore W4385482734C104317684 @default.
- W4385482734 hasConceptScore W4385482734C119857082 @default.
- W4385482734 hasConceptScore W4385482734C132525143 @default.
- W4385482734 hasConceptScore W4385482734C141231307 @default.
- W4385482734 hasConceptScore W4385482734C153180895 @default.
- W4385482734 hasConceptScore W4385482734C154945302 @default.
- W4385482734 hasConceptScore W4385482734C204321447 @default.
- W4385482734 hasConceptScore W4385482734C41008148 @default.
- W4385482734 hasConceptScore W4385482734C55493867 @default.
- W4385482734 hasConceptScore W4385482734C7386963 @default.
- W4385482734 hasConceptScore W4385482734C80444323 @default.
- W4385482734 hasConceptScore W4385482734C86803240 @default.
- W4385482734 hasLocation W43854827341 @default.
- W4385482734 hasOpenAccess W4385482734 @default.
- W4385482734 hasPrimaryLocation W43854827341 @default.
- W4385482734 hasRelatedWork W2961085424 @default.
- W4385482734 hasRelatedWork W3046775127 @default.
- W4385482734 hasRelatedWork W3170094116 @default.
- W4385482734 hasRelatedWork W4205958290 @default.
- W4385482734 hasRelatedWork W4285260836 @default.
- W4385482734 hasRelatedWork W4286629047 @default.
- W4385482734 hasRelatedWork W4306321456 @default.
- W4385482734 hasRelatedWork W4306674287 @default.
- W4385482734 hasRelatedWork W4386462264 @default.
- W4385482734 hasRelatedWork W4224009465 @default.
- W4385482734 isParatext "false" @default.
- W4385482734 isRetracted "false" @default.
- W4385482734 workType "article" @default.