Matches in SemOpenAlex for { <https://semopenalex.org/work/W3159551428> ?p ?o ?g. }
- W3159551428 endingPage "3053" @default.
- W3159551428 startingPage "3045" @default.
- W3159551428 abstract "We present a collaborative learning method called Mutual Contrastive Learning (MCL) for general visual representation learning. The core idea of MCL is to perform mutual interaction and transfer of contrastive distributions among a cohort of networks. A crucial component of MCL is Interactive Contrastive Learning (ICL). Compared with vanilla contrastive learning, ICL can aggregate cross-network embedding information and maximize the lower bound to the mutual information between two networks. This enables each network to learn extra contrastive knowledge from others, leading to better feature representations for visual recognition tasks. We emphasize that the resulting MCL is conceptually simple yet empirically powerful. It is a generic framework that can be applied to both supervised and self-supervised representation learning. Experimental results on image classification and transfer learning to object detection show that MCL can lead to consistent performance gains, demonstrating that MCL can guide the network to generate better feature representations. Code is available at https://github.com/winycg/MCL." @default.
- W3159551428 created "2021-05-10" @default.
- W3159551428 creator A5046293929 @default.
- W3159551428 creator A5055016576 @default.
- W3159551428 creator A5070277080 @default.
- W3159551428 creator A5085502749 @default.
- W3159551428 date "2022-06-28" @default.
- W3159551428 modified "2023-10-05" @default.
- W3159551428 title "Mutual Contrastive Learning for Visual Representation Learning" @default.
- W3159551428 cites W2031489346 @default.
- W3159551428 cites W2106053110 @default.
- W3159551428 cites W2108598243 @default.
- W3159551428 cites W2118858186 @default.
- W3159551428 cites W2138621090 @default.
- W3159551428 cites W2194775991 @default.
- W3159551428 cites W2321533354 @default.
- W3159551428 cites W2326925005 @default.
- W3159551428 cites W2565639579 @default.
- W3159551428 cites W2598634450 @default.
- W3159551428 cites W2601450892 @default.
- W3159551428 cites W2620998106 @default.
- W3159551428 cites W2798991696 @default.
- W3159551428 cites W2842511635 @default.
- W3159551428 cites W2921310091 @default.
- W3159551428 cites W2948638722 @default.
- W3159551428 cites W2949517790 @default.
- W3159551428 cites W2962758570 @default.
- W3159551428 cites W2963026686 @default.
- W3159551428 cites W2963341924 @default.
- W3159551428 cites W2963373431 @default.
- W3159551428 cites W2963420272 @default.
- W3159551428 cites W2963446712 @default.
- W3159551428 cites W2964137095 @default.
- W3159551428 cites W2964220233 @default.
- W3159551428 cites W2996970889 @default.
- W3159551428 cites W3018378048 @default.
- W3159551428 cites W3034235489 @default.
- W3159551428 cites W3034781633 @default.
- W3159551428 cites W3034978746 @default.
- W3159551428 cites W3035524453 @default.
- W3159551428 cites W3096621598 @default.
- W3159551428 cites W3099206234 @default.
- W3159551428 cites W3106974277 @default.
- W3159551428 cites W3107220607 @default.
- W3159551428 cites W3108411658 @default.
- W3159551428 cites W3118608800 @default.
- W3159551428 cites W639708223 @default.
- W3159551428 doi "https://doi.org/10.1609/aaai.v36i3.20211" @default.
- W3159551428 hasPublicationYear "2022" @default.
- W3159551428 type Work @default.
- W3159551428 sameAs 3159551428 @default.
- W3159551428 citedByCount "15" @default.
- W3159551428 countsByYear W31595514282022 @default.
- W3159551428 countsByYear W31595514282023 @default.
- W3159551428 crossrefType "journal-article" @default.
- W3159551428 hasAuthorship W3159551428A5046293929 @default.
- W3159551428 hasAuthorship W3159551428A5055016576 @default.
- W3159551428 hasAuthorship W3159551428A5070277080 @default.
- W3159551428 hasAuthorship W3159551428A5085502749 @default.
- W3159551428 hasBestOaLocation W31595514281 @default.
- W3159551428 hasConcept C119857082 @default.
- W3159551428 hasConcept C138885662 @default.
- W3159551428 hasConcept C150899416 @default.
- W3159551428 hasConcept C152139883 @default.
- W3159551428 hasConcept C153180895 @default.
- W3159551428 hasConcept C154945302 @default.
- W3159551428 hasConcept C17744445 @default.
- W3159551428 hasConcept C199539241 @default.
- W3159551428 hasConcept C204321447 @default.
- W3159551428 hasConcept C2776359362 @default.
- W3159551428 hasConcept C2776401178 @default.
- W3159551428 hasConcept C2781238097 @default.
- W3159551428 hasConcept C41008148 @default.
- W3159551428 hasConcept C41608201 @default.
- W3159551428 hasConcept C41895202 @default.
- W3159551428 hasConcept C59404180 @default.
- W3159551428 hasConcept C94625758 @default.
- W3159551428 hasConceptScore W3159551428C119857082 @default.
- W3159551428 hasConceptScore W3159551428C138885662 @default.
- W3159551428 hasConceptScore W3159551428C150899416 @default.
- W3159551428 hasConceptScore W3159551428C152139883 @default.
- W3159551428 hasConceptScore W3159551428C153180895 @default.
- W3159551428 hasConceptScore W3159551428C154945302 @default.
- W3159551428 hasConceptScore W3159551428C17744445 @default.
- W3159551428 hasConceptScore W3159551428C199539241 @default.
- W3159551428 hasConceptScore W3159551428C204321447 @default.
- W3159551428 hasConceptScore W3159551428C2776359362 @default.
- W3159551428 hasConceptScore W3159551428C2776401178 @default.
- W3159551428 hasConceptScore W3159551428C2781238097 @default.
- W3159551428 hasConceptScore W3159551428C41008148 @default.
- W3159551428 hasConceptScore W3159551428C41608201 @default.
- W3159551428 hasConceptScore W3159551428C41895202 @default.
- W3159551428 hasConceptScore W3159551428C59404180 @default.
- W3159551428 hasConceptScore W3159551428C94625758 @default.
- W3159551428 hasIssue "3" @default.
- W3159551428 hasLocation W31595514281 @default.
- W3159551428 hasLocation W31595514282 @default.
- W3159551428 hasOpenAccess W3159551428 @default.