FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386526950> ?p ?o ?g. }

• W4386526950 abstract "Abstract Transferring knowledge from pre-trained deep models for downstream tasks, particularly with limited labeled samples, is a fundamental problem in computer vision research. Recent advances in large-scale, task-agnostic vision-language pre-trained models, which are learned with billions of samples, have shed new light on this problem. In this study, we investigate how to efficiently transfer aligned visual and textual knowledge for downstream visual recognition tasks. We first revisit the role of the linear classifier in the vanilla transfer learning framework, and then propose a new paradigm where the parameters of the classifier are initialized with semantic targets from the textual encoder and remain fixed during optimization. To provide a comparison, we also initialize the classifier with knowledge from various resources. In the empirical study, we demonstrate that our paradigm improves the performance and training speed of transfer learning tasks. With only minor modifications, our approach proves effective across 17 visual datasets that span three different data domains: image, video, and 3D point cloud." @default.
• W4386526950 created "2023-09-08" @default.
• W4386526950 creator A5015819673 @default.
• W4386526950 creator A5016846734 @default.
• W4386526950 creator A5071604338 @default.
• W4386526950 creator A5075880303 @default.
• W4386526950 creator A5087818121 @default.
• W4386526950 date "2023-09-07" @default.
• W4386526950 modified "2023-10-16" @default.
• W4386526950 title "Transferring Vision-Language Models for Visual Recognition: A Classifier Perspective" @default.
• W4386526950 cites W12634471 @default.
• W4386526950 cites W1927052826 @default.
• W4386526950 cites W1977295328 @default.
• W4386526950 cites W2017814585 @default.
• W4386526950 cites W2047643928 @default.
• W4386526950 cites W2108598243 @default.
• W4386526950 cites W2108950639 @default.
• W4386526950 cites W2126579184 @default.
• W4386526950 cites W2138011018 @default.
• W4386526950 cites W2155904486 @default.
• W4386526950 cites W2194775991 @default.
• W4386526950 cites W2337252826 @default.
• W4386526950 cites W2533598788 @default.
• W4386526950 cites W2770804203 @default.
• W4386526950 cites W2883429621 @default.
• W4386526950 cites W2887280559 @default.
• W4386526950 cites W2904378456 @default.
• W4386526950 cites W2955874753 @default.
• W4386526950 cites W2962843773 @default.
• W4386526950 cites W2963091558 @default.
• W4386526950 cites W2963155035 @default.
• W4386526950 cites W2963524571 @default.
• W4386526950 cites W2963645879 @default.
• W4386526950 cites W2963689837 @default.
• W4386526950 cites W2963820951 @default.
• W4386526950 cites W2964194231 @default.
• W4386526950 cites W2980037812 @default.
• W4386526950 cites W2981385151 @default.
• W4386526950 cites W2984287396 @default.
• W4386526950 cites W2990152177 @default.
• W4386526950 cites W2990503944 @default.
• W4386526950 cites W2993751684 @default.
• W4386526950 cites W2996901793 @default.
• W4386526950 cites W3010010212 @default.
• W4386526950 cites W3034572008 @default.
• W4386526950 cites W3034658206 @default.
• W4386526950 cites W3035254087 @default.
• W4386526950 cites W3035524453 @default.
• W4386526950 cites W3041133507 @default.
• W4386526950 cites W3138516171 @default.
• W4386526950 cites W3145450063 @default.
• W4386526950 cites W3172942063 @default.
• W4386526950 cites W3174568846 @default.
• W4386526950 cites W3175528717 @default.
• W4386526950 cites W3176125528 @default.
• W4386526950 cites W3207340843 @default.
• W4386526950 cites W4214612132 @default.
• W4386526950 cites W4214614183 @default.
• W4386526950 cites W4214746887 @default.
• W4386526950 cites W4225414521 @default.
• W4386526950 cites W4226058394 @default.
• W4386526950 cites W4285606530 @default.
• W4386526950 cites W4312254032 @default.
• W4386526950 cites W4312266966 @default.
• W4386526950 cites W4312302951 @default.
• W4386526950 cites W4312310776 @default.
• W4386526950 cites W4312420092 @default.
• W4386526950 cites W4312480274 @default.
• W4386526950 cites W4312558481 @default.
• W4386526950 cites W4312560592 @default.
• W4386526950 cites W4312614039 @default.
• W4386526950 cites W4312658081 @default.
• W4386526950 cites W4312818263 @default.
• W4386526950 cites W4313136445 @default.
• W4386526950 cites W4313156423 @default.
• W4386526950 doi "https://doi.org/10.1007/s11263-023-01876-w" @default.
• W4386526950 hasPublicationYear "2023" @default.
• W4386526950 type Work @default.
• W4386526950 citedByCount "0" @default.
• W4386526950 crossrefType "journal-article" @default.
• W4386526950 hasAuthorship W4386526950A5015819673 @default.
• W4386526950 hasAuthorship W4386526950A5016846734 @default.
• W4386526950 hasAuthorship W4386526950A5071604338 @default.
• W4386526950 hasAuthorship W4386526950A5075880303 @default.
• W4386526950 hasAuthorship W4386526950A5087818121 @default.
• W4386526950 hasBestOaLocation W43865269501 @default.
• W4386526950 hasConcept C111919701 @default.
• W4386526950 hasConcept C118505674 @default.
• W4386526950 hasConcept C119857082 @default.
• W4386526950 hasConcept C150899416 @default.
• W4386526950 hasConcept C153180895 @default.
• W4386526950 hasConcept C154945302 @default.
• W4386526950 hasConcept C204321447 @default.
• W4386526950 hasConcept C41008148 @default.
• W4386526950 hasConcept C95623464 @default.
• W4386526950 hasConceptScore W4386526950C111919701 @default.
• W4386526950 hasConceptScore W4386526950C118505674 @default.
• W4386526950 hasConceptScore W4386526950C119857082 @default.
• W4386526950 hasConceptScore W4386526950C150899416 @default.
• W4386526950 hasConceptScore W4386526950C153180895 @default.

• next