Matches in SemOpenAlex for { <https://semopenalex.org/work/W4214868967> ?p ?o ?g. }
- W4214868967 endingPage "287" @default.
- W4214868967 startingPage "279" @default.
- W4214868967 abstract "Molecular Machine Learning (ML) bears promise for efficient molecule property prediction and drug discovery. However, labeled molecule data can be expensive and time-consuming to acquire. Due to the limited labeled data, it is a great challenge for supervised-learning ML models to generalize to the giant chemical space. In this work, we present MolCLR: Molecular Contrastive Learning of Representations via Graph Neural Networks (GNNs), a self-supervised learning framework that leverages large unlabeled data (~10M unique molecules). In MolCLR pre-training, we build molecule graphs and develop GNN encoders to learn differentiable representations. Three molecule graph augmentations are proposed: atom masking, bond deletion, and subgraph removal. A contrastive estimator maximizes the agreement of augmentations from the same molecule while minimizing the agreement of different molecules. Experiments show that our contrastive learning framework significantly improves the performance of GNNs on various molecular property benchmarks including both classification and regression tasks. Benefiting from pre-training on the large unlabeled database, MolCLR even achieves state-of-the-art on several challenging benchmarks after fine-tuning. Additionally, further investigations demonstrate that MolCLR learns to embed molecules into representations that can distinguish chemically reasonable molecular similarities." @default.
- W4214868967 created "2022-03-05" @default.
- W4214868967 creator A5008745801 @default.
- W4214868967 creator A5037075675 @default.
- W4214868967 creator A5043876952 @default.
- W4214868967 creator A5073340077 @default.
- W4214868967 date "2022-03-03" @default.
- W4214868967 modified "2023-10-11" @default.
- W4214868967 title "Molecular contrastive learning of representations via graph neural networks" @default.
- W4214868967 cites W1757990252 @default.
- W4214868967 cites W1975147762 @default.
- W4214868967 cites W1988037271 @default.
- W4214868967 cites W1999798000 @default.
- W4214868967 cites W2017254234 @default.
- W4214868967 cites W2025948031 @default.
- W4214868967 cites W2029413789 @default.
- W4214868967 cites W2096541451 @default.
- W4214868967 cites W2558748708 @default.
- W4214868967 cites W2565684601 @default.
- W4214868967 cites W2594183968 @default.
- W4214868967 cites W2749279690 @default.
- W4214868967 cites W2765224015 @default.
- W4214868967 cites W2778051509 @default.
- W4214868967 cites W2792137452 @default.
- W4214868967 cites W2895884529 @default.
- W4214868967 cites W2899070097 @default.
- W4214868967 cites W2919115771 @default.
- W4214868967 cites W2937307539 @default.
- W4214868967 cites W2953128081 @default.
- W4214868967 cites W2962876364 @default.
- W4214868967 cites W2963396480 @default.
- W4214868967 cites W2966357564 @default.
- W4214868967 cites W2973114758 @default.
- W4214868967 cites W2979481854 @default.
- W4214868967 cites W3035524453 @default.
- W4214868967 cites W3045928028 @default.
- W4214868967 cites W3049517558 @default.
- W4214868967 cites W3087318293 @default.
- W4214868967 cites W3098269892 @default.
- W4214868967 cites W3098370560 @default.
- W4214868967 cites W3102659967 @default.
- W4214868967 cites W3103092523 @default.
- W4214868967 cites W3106310231 @default.
- W4214868967 cites W3125600935 @default.
- W4214868967 cites W3135622340 @default.
- W4214868967 cites W3156636935 @default.
- W4214868967 cites W4242320552 @default.
- W4214868967 doi "https://doi.org/10.1038/s42256-022-00447-x" @default.
- W4214868967 hasPublicationYear "2022" @default.
- W4214868967 type Work @default.
- W4214868967 citedByCount "99" @default.
- W4214868967 countsByYear W42148689672022 @default.
- W4214868967 countsByYear W42148689672023 @default.
- W4214868967 crossrefType "journal-article" @default.
- W4214868967 hasAuthorship W4214868967A5008745801 @default.
- W4214868967 hasAuthorship W4214868967A5037075675 @default.
- W4214868967 hasAuthorship W4214868967A5043876952 @default.
- W4214868967 hasAuthorship W4214868967A5073340077 @default.
- W4214868967 hasBestOaLocation W42148689672 @default.
- W4214868967 hasConcept C119857082 @default.
- W4214868967 hasConcept C132525143 @default.
- W4214868967 hasConcept C147597530 @default.
- W4214868967 hasConcept C153180895 @default.
- W4214868967 hasConcept C154945302 @default.
- W4214868967 hasConcept C185592680 @default.
- W4214868967 hasConcept C204321447 @default.
- W4214868967 hasConcept C2780022179 @default.
- W4214868967 hasConcept C41008148 @default.
- W4214868967 hasConcept C50644808 @default.
- W4214868967 hasConcept C55493867 @default.
- W4214868967 hasConcept C68762167 @default.
- W4214868967 hasConcept C74187038 @default.
- W4214868967 hasConcept C80444323 @default.
- W4214868967 hasConcept C99726746 @default.
- W4214868967 hasConceptScore W4214868967C119857082 @default.
- W4214868967 hasConceptScore W4214868967C132525143 @default.
- W4214868967 hasConceptScore W4214868967C147597530 @default.
- W4214868967 hasConceptScore W4214868967C153180895 @default.
- W4214868967 hasConceptScore W4214868967C154945302 @default.
- W4214868967 hasConceptScore W4214868967C185592680 @default.
- W4214868967 hasConceptScore W4214868967C204321447 @default.
- W4214868967 hasConceptScore W4214868967C2780022179 @default.
- W4214868967 hasConceptScore W4214868967C41008148 @default.
- W4214868967 hasConceptScore W4214868967C50644808 @default.
- W4214868967 hasConceptScore W4214868967C55493867 @default.
- W4214868967 hasConceptScore W4214868967C68762167 @default.
- W4214868967 hasConceptScore W4214868967C74187038 @default.
- W4214868967 hasConceptScore W4214868967C80444323 @default.
- W4214868967 hasConceptScore W4214868967C99726746 @default.
- W4214868967 hasIssue "3" @default.
- W4214868967 hasLocation W42148689671 @default.
- W4214868967 hasLocation W42148689672 @default.
- W4214868967 hasLocation W42148689673 @default.
- W4214868967 hasOpenAccess W4214868967 @default.
- W4214868967 hasPrimaryLocation W42148689671 @default.
- W4214868967 hasRelatedWork W2160244398 @default.
- W4214868967 hasRelatedWork W2601867172 @default.
- W4214868967 hasRelatedWork W2902857455 @default.