FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3206658641> ?p ?o ?g. }

• W3206658641 abstract "Domain adaptation (DA) paves the way for label annotation and dataset bias issues by the knowledge transfer from a label-rich source domain to a related but unlabeled target domain. A mainstream of DA methods is to align the feature distributions of the two domains. However, the majority of them focus on the entire image features where irrelevant semantic information, e.g., the messy background, is inevitably embedded. Enforcing feature alignments in such case will negatively influence the correct matching of objects and consequently lead to the semantically negative transfer due to the confusion of irrelevant semantics. To tackle this issue, we propose Semantic Concentration for Domain Adaptation (SCDA), which encourages the model to concentrate on the most principal features via the pair-wise adversarial alignment of prediction distributions. Specifically, we train the classifier to class-wisely maximize the prediction distribution divergence of each sample pair, which enables the model to find the region with large differences among the same class of samples. Meanwhile, the feature extractor attempts to minimize that discrepancy, which suppresses the features of dissimilar regions among the same class of samples and accentuates the features of principal parts. As a general method, SCDA can be easily integrated into various DA methods as a regularizer to further boost their performance. Extensive experiments on the cross-domain benchmarks show the efficacy of SCDA." @default.
• W3206658641 created "2021-10-25" @default.
• W3206658641 creator A5000432967 @default.
• W3206658641 creator A5001817133 @default.
• W3206658641 creator A5028293439 @default.
• W3206658641 creator A5034541026 @default.
• W3206658641 creator A5054547171 @default.
• W3206658641 creator A5061084605 @default.
• W3206658641 creator A5086988988 @default.
• W3206658641 date "2021-08-12" @default.
• W3206658641 modified "2023-09-25" @default.
• W3206658641 title "Semantic Concentration for Domain Adaptation" @default.
• W3206658641 cites W1722318740 @default.
• W3206658641 cites W1821462560 @default.
• W3206658641 cites W2099471712 @default.
• W3206658641 cites W2104094955 @default.
• W3206658641 cites W2117539524 @default.
• W3206658641 cites W2125865219 @default.
• W3206658641 cites W2131953535 @default.
• W3206658641 cites W2145494108 @default.
• W3206658641 cites W2147527908 @default.
• W3206658641 cites W2155541015 @default.
• W3206658641 cites W2159291411 @default.
• W3206658641 cites W2163605009 @default.
• W3206658641 cites W2165698076 @default.
• W3206658641 cites W2187089797 @default.
• W3206658641 cites W2194775991 @default.
• W3206658641 cites W2295107390 @default.
• W3206658641 cites W2412782625 @default.
• W3206658641 cites W2613718673 @default.
• W3206658641 cites W2627183927 @default.
• W3206658641 cites W2767179670 @default.
• W3206658641 cites W2785416514 @default.
• W3206658641 cites W2786559811 @default.
• W3206658641 cites W2795155917 @default.
• W3206658641 cites W2904170036 @default.
• W3206658641 cites W2904706552 @default.
• W3206658641 cites W2925229117 @default.
• W3206658641 cites W2945164914 @default.
• W3206658641 cites W2945328857 @default.
• W3206658641 cites W2946444988 @default.
• W3206658641 cites W2949987290 @default.
• W3206658641 cites W2950561908 @default.
• W3206658641 cites W2962687275 @default.
• W3206658641 cites W2962858109 @default.
• W3206658641 cites W2963094258 @default.
• W3206658641 cites W2963403868 @default.
• W3206658641 cites W2963826681 @default.
• W3206658641 cites W2964139811 @default.
• W3206658641 cites W2964278684 @default.
• W3206658641 cites W2964288524 @default.
• W3206658641 cites W2970971581 @default.
• W3206658641 cites W2971201880 @default.
• W3206658641 cites W2981720610 @default.
• W3206658641 cites W2981885563 @default.
• W3206658641 cites W2987861506 @default.
• W3206658641 cites W2997739323 @default.
• W3206658641 cites W3004810941 @default.
• W3206658641 cites W3014059139 @default.
• W3206658641 cites W3034695001 @default.
• W3206658641 cites W3035456997 @default.
• W3206658641 cites W3087144073 @default.
• W3206658641 cites W3109093849 @default.
• W3206658641 cites W3112708475 @default.
• W3206658641 doi "https://doi.org/10.48550/arxiv.2108.05720" @default.
• W3206658641 hasPublicationYear "2021" @default.
• W3206658641 type Work @default.
• W3206658641 sameAs 3206658641 @default.
• W3206658641 citedByCount "0" @default.
• W3206658641 crossrefType "posted-content" @default.
• W3206658641 hasAuthorship W3206658641A5000432967 @default.
• W3206658641 hasAuthorship W3206658641A5001817133 @default.
• W3206658641 hasAuthorship W3206658641A5028293439 @default.
• W3206658641 hasAuthorship W3206658641A5034541026 @default.
• W3206658641 hasAuthorship W3206658641A5054547171 @default.
• W3206658641 hasAuthorship W3206658641A5061084605 @default.
• W3206658641 hasAuthorship W3206658641A5086988988 @default.
• W3206658641 hasBestOaLocation W32066586411 @default.
• W3206658641 hasConcept C105795698 @default.
• W3206658641 hasConcept C117978034 @default.
• W3206658641 hasConcept C119857082 @default.
• W3206658641 hasConcept C120665830 @default.
• W3206658641 hasConcept C121332964 @default.
• W3206658641 hasConcept C127413603 @default.
• W3206658641 hasConcept C134306372 @default.
• W3206658641 hasConcept C138885662 @default.
• W3206658641 hasConcept C153180895 @default.
• W3206658641 hasConcept C154945302 @default.
• W3206658641 hasConcept C165064840 @default.
• W3206658641 hasConcept C184337299 @default.
• W3206658641 hasConcept C192209626 @default.
• W3206658641 hasConcept C199360897 @default.
• W3206658641 hasConcept C204321447 @default.
• W3206658641 hasConcept C21880701 @default.
• W3206658641 hasConcept C2776321320 @default.
• W3206658641 hasConcept C2776401178 @default.
• W3206658641 hasConcept C2776434776 @default.
• W3206658641 hasConcept C33923547 @default.
• W3206658641 hasConcept C36503486 @default.
• W3206658641 hasConcept C41008148 @default.

• next