Matches in SemOpenAlex for { <https://semopenalex.org/work/W2985836869> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W2985836869 endingPage "2438" @default.
- W2985836869 startingPage "2424" @default.
- W2985836869 abstract "Subspace learning and reconstruction have been widely explored in recent transfer learning work. Generally, a specially designed projection and reconstruction transfer functions bridging multiple domains for heterogeneous knowledge sharing are wanted. However, we argue that the existing subspace reconstruction based domain adaptation algorithms neglect the class prior, such that the learned transfer function is biased, especially when data scarcity of some class is encountered. Different from those previous methods, in this article, we propose a novel class-wise reconstruction-based adaptation method called Class-specific Reconstruction Transfer Learning (CRTL), which optimizes a well modeled transfer loss function by fully exploiting intra-class dependency and inter-class independency. The merits of the CRTL are three-fold. 1) Using a class-specific reconstruction matrix to align the source domain with the target domain fully exploits the class prior in modeling the domain distribution consistency, which benefits the cross-domain classification. 2) Furthermore, to keep the intrinsic relationship between data and labels after feature augmentation, a projected Hilbert-Schmidt Independence Criterion (pHSIC), that measures the dependency between data and label, is first proposed in transfer learning community by mapping the data from raw space to RKHS. 3) In addition, by imposing low-rank and sparse constraints on the class-specific reconstruction coefficient matrix, the global and local data structure that contributes to domain correlation can be effectively preserved. Extensive experiments on challenging benchmark datasets demonstrate the superiority of the proposed method over state-of-the-art representation-based domain adaptation methods. The demo code is available in https://github.com/wangshanshanCQU/CRTL." @default.
- W2985836869 created "2019-11-22" @default.
- W2985836869 creator A5003381889 @default.
- W2985836869 creator A5018318136 @default.
- W2985836869 creator A5048088901 @default.
- W2985836869 creator A5090034491 @default.
- W2985836869 date "2020-01-01" @default.
- W2985836869 modified "2023-10-06" @default.
- W2985836869 title "Class-Specific Reconstruction Transfer Learning for Visual Recognition Across Domains" @default.
- W2985836869 cites W1537211966 @default.
- W2985836869 cites W1638081485 @default.
- W2985836869 cites W1722318740 @default.
- W2985836869 cites W1916649859 @default.
- W2985836869 cites W1972420097 @default.
- W2985836869 cites W1978920452 @default.
- W2985836869 cites W1985967702 @default.
- W2985836869 cites W1993962865 @default.
- W2985836869 cites W1995349519 @default.
- W2985836869 cites W1997201895 @default.
- W2985836869 cites W2002329661 @default.
- W2985836869 cites W2009668020 @default.
- W2985836869 cites W2014277868 @default.
- W2985836869 cites W2046385669 @default.
- W2985836869 cites W2059477850 @default.
- W2985836869 cites W2075728230 @default.
- W2985836869 cites W2090923791 @default.
- W2985836869 cites W2096943734 @default.
- W2985836869 cites W2103972604 @default.
- W2985836869 cites W2104068492 @default.
- W2985836869 cites W2128053425 @default.
- W2985836869 cites W2153963799 @default.
- W2985836869 cites W2157032359 @default.
- W2985836869 cites W2157785665 @default.
- W2985836869 cites W2161381512 @default.
- W2985836869 cites W2211868589 @default.
- W2985836869 cites W2214409633 @default.
- W2985836869 cites W2240559667 @default.
- W2985836869 cites W2283717164 @default.
- W2985836869 cites W2466618734 @default.
- W2985836869 cites W2517537544 @default.
- W2985836869 cites W2593768305 @default.
- W2985836869 cites W2743316710 @default.
- W2985836869 cites W2763549966 @default.
- W2985836869 cites W2770507894 @default.
- W2985836869 cites W2770824758 @default.
- W2985836869 cites W2923769473 @default.
- W2985836869 cites W4232262564 @default.
- W2985836869 doi "https://doi.org/10.1109/tip.2019.2948480" @default.
- W2985836869 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31714223" @default.
- W2985836869 hasPublicationYear "2020" @default.
- W2985836869 type Work @default.
- W2985836869 sameAs 2985836869 @default.
- W2985836869 citedByCount "64" @default.
- W2985836869 countsByYear W29858368692020 @default.
- W2985836869 countsByYear W29858368692021 @default.
- W2985836869 countsByYear W29858368692022 @default.
- W2985836869 countsByYear W29858368692023 @default.
- W2985836869 crossrefType "journal-article" @default.
- W2985836869 hasAuthorship W2985836869A5003381889 @default.
- W2985836869 hasAuthorship W2985836869A5018318136 @default.
- W2985836869 hasAuthorship W2985836869A5048088901 @default.
- W2985836869 hasAuthorship W2985836869A5090034491 @default.
- W2985836869 hasConcept C119857082 @default.
- W2985836869 hasConcept C150899416 @default.
- W2985836869 hasConcept C153180895 @default.
- W2985836869 hasConcept C154945302 @default.
- W2985836869 hasConcept C32834561 @default.
- W2985836869 hasConcept C41008148 @default.
- W2985836869 hasConceptScore W2985836869C119857082 @default.
- W2985836869 hasConceptScore W2985836869C150899416 @default.
- W2985836869 hasConceptScore W2985836869C153180895 @default.
- W2985836869 hasConceptScore W2985836869C154945302 @default.
- W2985836869 hasConceptScore W2985836869C32834561 @default.
- W2985836869 hasConceptScore W2985836869C41008148 @default.
- W2985836869 hasFunder F4320321001 @default.
- W2985836869 hasLocation W29858368691 @default.
- W2985836869 hasLocation W29858368692 @default.
- W2985836869 hasOpenAccess W2985836869 @default.
- W2985836869 hasPrimaryLocation W29858368691 @default.
- W2985836869 hasRelatedWork W2321141263 @default.
- W2985836869 hasRelatedWork W2543161807 @default.
- W2985836869 hasRelatedWork W2960456850 @default.
- W2985836869 hasRelatedWork W3021430260 @default.
- W2985836869 hasRelatedWork W4281645081 @default.
- W2985836869 hasRelatedWork W4308262314 @default.
- W2985836869 hasRelatedWork W4312200629 @default.
- W2985836869 hasRelatedWork W4317565044 @default.
- W2985836869 hasRelatedWork W4382286161 @default.
- W2985836869 hasRelatedWork W4386213806 @default.
- W2985836869 hasVolume "29" @default.
- W2985836869 isParatext "false" @default.
- W2985836869 isRetracted "false" @default.
- W2985836869 magId "2985836869" @default.
- W2985836869 workType "article" @default.