Matches in SemOpenAlex for { <https://semopenalex.org/work/W4382655160> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4382655160 endingPage "126491" @default.
- W4382655160 startingPage "126491" @default.
- W4382655160 abstract "Contrastive learning as an effective representation learning technique has attracted tremendous attention due to its general success in downstream tasks. However, the theoretical explanations and quantitative experimental analyses of its generalization ability are still limited. These issues are pivotal yet challenging for improving both the interpretability and performance of contrastive learning. To address these issues, we first re-examine the least squares bias-variance decomposition and successfully derive GCL, a novel bias-variance decomposition with two optional generalized biases and one generalized variance. GCL is shown to be extendable to common contrastive learning models so that it can be utilized as a unified contrastive learning framework. Meanwhile, a surprising finding that the gradient descent of contrastive loss concerning feature representation is closely related to the message passing mechanism (graph convolution) of Graph Neural Networks (GNNs). The contrastive learning model called GCP is then proposed as a convincing implementation of GCL. GCP has a pure MLP-based structure and employs a conventional cross-entropy to reduce the bias between predictions and ground truth labels, two optional contrastive losses to optimize the variance of the model. Finally, extensive experiments demonstrate that the two biases proposed by GCL have their own merits; GCP achieves comparable or even better performance than GNNs in a more efficient and robust manner, its bias and variance meet the bias-variance tradeoff to some extent." @default.
- W4382655160 created "2023-07-01" @default.
- W4382655160 creator A5003642180 @default.
- W4382655160 creator A5044029807 @default.
- W4382655160 creator A5052097137 @default.
- W4382655160 creator A5064824079 @default.
- W4382655160 creator A5070077478 @default.
- W4382655160 creator A5077075529 @default.
- W4382655160 date "2023-09-01" @default.
- W4382655160 modified "2023-10-17" @default.
- W4382655160 title "GCL: Contrastive learning instead of graph convolution for node classification" @default.
- W4382655160 cites W2048305092 @default.
- W4382655160 cites W2094051685 @default.
- W4382655160 cites W2169502938 @default.
- W4382655160 cites W2923764619 @default.
- W4382655160 cites W2964015721 @default.
- W4382655160 cites W3049694790 @default.
- W4382655160 cites W3111350549 @default.
- W4382655160 cites W4210475840 @default.
- W4382655160 cites W4280531754 @default.
- W4382655160 cites W4280582438 @default.
- W4382655160 cites W4328031094 @default.
- W4382655160 doi "https://doi.org/10.1016/j.neucom.2023.126491" @default.
- W4382655160 hasPublicationYear "2023" @default.
- W4382655160 type Work @default.
- W4382655160 citedByCount "0" @default.
- W4382655160 crossrefType "journal-article" @default.
- W4382655160 hasAuthorship W4382655160A5003642180 @default.
- W4382655160 hasAuthorship W4382655160A5044029807 @default.
- W4382655160 hasAuthorship W4382655160A5052097137 @default.
- W4382655160 hasAuthorship W4382655160A5064824079 @default.
- W4382655160 hasAuthorship W4382655160A5070077478 @default.
- W4382655160 hasAuthorship W4382655160A5077075529 @default.
- W4382655160 hasConcept C11413529 @default.
- W4382655160 hasConcept C119857082 @default.
- W4382655160 hasConcept C121955636 @default.
- W4382655160 hasConcept C132525143 @default.
- W4382655160 hasConcept C134306372 @default.
- W4382655160 hasConcept C144133560 @default.
- W4382655160 hasConcept C153180895 @default.
- W4382655160 hasConcept C154945302 @default.
- W4382655160 hasConcept C177148314 @default.
- W4382655160 hasConcept C196083921 @default.
- W4382655160 hasConcept C2781067378 @default.
- W4382655160 hasConcept C33923547 @default.
- W4382655160 hasConcept C41008148 @default.
- W4382655160 hasConcept C45347329 @default.
- W4382655160 hasConcept C50644808 @default.
- W4382655160 hasConcept C59404180 @default.
- W4382655160 hasConcept C80444323 @default.
- W4382655160 hasConceptScore W4382655160C11413529 @default.
- W4382655160 hasConceptScore W4382655160C119857082 @default.
- W4382655160 hasConceptScore W4382655160C121955636 @default.
- W4382655160 hasConceptScore W4382655160C132525143 @default.
- W4382655160 hasConceptScore W4382655160C134306372 @default.
- W4382655160 hasConceptScore W4382655160C144133560 @default.
- W4382655160 hasConceptScore W4382655160C153180895 @default.
- W4382655160 hasConceptScore W4382655160C154945302 @default.
- W4382655160 hasConceptScore W4382655160C177148314 @default.
- W4382655160 hasConceptScore W4382655160C196083921 @default.
- W4382655160 hasConceptScore W4382655160C2781067378 @default.
- W4382655160 hasConceptScore W4382655160C33923547 @default.
- W4382655160 hasConceptScore W4382655160C41008148 @default.
- W4382655160 hasConceptScore W4382655160C45347329 @default.
- W4382655160 hasConceptScore W4382655160C50644808 @default.
- W4382655160 hasConceptScore W4382655160C59404180 @default.
- W4382655160 hasConceptScore W4382655160C80444323 @default.
- W4382655160 hasLocation W43826551601 @default.
- W4382655160 hasOpenAccess W4382655160 @default.
- W4382655160 hasPrimaryLocation W43826551601 @default.
- W4382655160 hasRelatedWork W3006943036 @default.
- W4382655160 hasRelatedWork W3080532707 @default.
- W4382655160 hasRelatedWork W4200511449 @default.
- W4382655160 hasRelatedWork W4206534706 @default.
- W4382655160 hasRelatedWork W4229079080 @default.
- W4382655160 hasRelatedWork W4287995534 @default.
- W4382655160 hasRelatedWork W4299487748 @default.
- W4382655160 hasRelatedWork W4385957992 @default.
- W4382655160 hasRelatedWork W4385965371 @default.
- W4382655160 hasRelatedWork W4386025632 @default.
- W4382655160 hasVolume "551" @default.
- W4382655160 isParatext "false" @default.
- W4382655160 isRetracted "false" @default.
- W4382655160 workType "article" @default.