FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2954400107> ?p ?o ?g. }

• W2954400107 endingPage "3010" @default.
• W2954400107 startingPage "2996" @default.
• W2954400107 abstract "With only coarse labels, weakly supervised learning typically uses top-down attention maps generated by back-propagating gradients as priors for tasks such as object localization and semantic segmentation. While these attention maps are intuitive and informative explanations of deep neural network, there is no effective mechanism to manipulate the network attention during learning process. In this paper, we address three shortcomings of previous approaches in modeling such attention maps in one common framework. First, we make attention maps a natural and explicit component in the training pipeline such that they are end-to-end trainable. Moreover, we provide self-guidance directly on these maps by exploring supervision from the network itself to improve them towards specific target tasks. Lastly, we proposed a design to seamlessly bridge the gap between using weak and extra supervision if available. Despite its simplicity, experiments on the semantic segmentation task demonstrate the effectiveness of our methods. Besides, the proposed framework provides a way not only explaining the focus of the learner but also feeding back with direct guidance towards specific tasks. Under mild assumptions our method can also be understood as a plug-in to existing convolutional neural networks to improve their generalization performance." @default.
• W2954400107 created "2019-07-12" @default.
• W2954400107 creator A5003798053 @default.
• W2954400107 creator A5005819096 @default.
• W2954400107 creator A5022886524 @default.
• W2954400107 creator A5027023070 @default.
• W2954400107 creator A5063443405 @default.
• W2954400107 date "2020-12-01" @default.
• W2954400107 modified "2023-10-16" @default.
• W2954400107 title "Guided Attention Inference Network" @default.
• W2954400107 cites W1495267108 @default.
• W2954400107 cites W1783315696 @default.
• W2954400107 cites W1849277567 @default.
• W2954400107 cites W1903029394 @default.
• W2954400107 cites W1945608308 @default.
• W2954400107 cites W1991367009 @default.
• W2954400107 cites W1994488211 @default.
• W2954400107 cites W2031342017 @default.
• W2954400107 cites W2031489346 @default.
• W2954400107 cites W2108598243 @default.
• W2954400107 cites W2133515615 @default.
• W2954400107 cites W2144794286 @default.
• W2954400107 cites W2155893237 @default.
• W2954400107 cites W2194775991 @default.
• W2954400107 cites W2221625691 @default.
• W2954400107 cites W2257483379 @default.
• W2954400107 cites W2295107390 @default.
• W2954400107 cites W2306289963 @default.
• W2954400107 cites W2412782625 @default.
• W2954400107 cites W2461475918 @default.
• W2954400107 cites W2503388974 @default.
• W2954400107 cites W2516803306 @default.
• W2954400107 cites W2519610629 @default.
• W2954400107 cites W2520746254 @default.
• W2954400107 cites W2559348937 @default.
• W2954400107 cites W2600144439 @default.
• W2954400107 cites W2606129492 @default.
• W2954400107 cites W2739450375 @default.
• W2954400107 cites W2740668812 @default.
• W2954400107 cites W2798376494 @default.
• W2954400107 cites W2866634454 @default.
• W2954400107 cites W2883554151 @default.
• W2954400107 cites W2886996240 @default.
• W2954400107 cites W2962758679 @default.
• W2954400107 cites W2962858109 @default.
• W2954400107 cites W2963045696 @default.
• W2954400107 cites W2963311325 @default.
• W2954400107 cites W2963371637 @default.
• W2954400107 cites W2963409068 @default.
• W2954400107 cites W2963446712 @default.
• W2954400107 cites W2963606198 @default.
• W2954400107 cites W2963670239 @default.
• W2954400107 cites W2963697527 @default.
• W2954400107 cites W2964101377 @default.
• W2954400107 cites W2964140013 @default.
• W2954400107 cites W2964274719 @default.
• W2954400107 cites W2972313371 @default.
• W2954400107 cites W611457968 @default.
• W2954400107 doi "https://doi.org/10.1109/tpami.2019.2921543" @default.
• W2954400107 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31180839" @default.
• W2954400107 hasPublicationYear "2020" @default.
• W2954400107 type Work @default.
• W2954400107 sameAs 2954400107 @default.
• W2954400107 citedByCount "35" @default.
• W2954400107 countsByYear W29544001072019 @default.
• W2954400107 countsByYear W29544001072020 @default.
• W2954400107 countsByYear W29544001072021 @default.
• W2954400107 countsByYear W29544001072022 @default.
• W2954400107 countsByYear W29544001072023 @default.
• W2954400107 crossrefType "journal-article" @default.
• W2954400107 hasAuthorship W2954400107A5003798053 @default.
• W2954400107 hasAuthorship W2954400107A5005819096 @default.
• W2954400107 hasAuthorship W2954400107A5022886524 @default.
• W2954400107 hasAuthorship W2954400107A5027023070 @default.
• W2954400107 hasAuthorship W2954400107A5063443405 @default.
• W2954400107 hasBestOaLocation W29544001071 @default.
• W2954400107 hasConcept C119857082 @default.
• W2954400107 hasConcept C154945302 @default.
• W2954400107 hasConcept C2776214188 @default.
• W2954400107 hasConcept C41008148 @default.
• W2954400107 hasConceptScore W2954400107C119857082 @default.
• W2954400107 hasConceptScore W2954400107C154945302 @default.
• W2954400107 hasConceptScore W2954400107C2776214188 @default.
• W2954400107 hasConceptScore W2954400107C41008148 @default.
• W2954400107 hasFunder F4320306076 @default.
• W2954400107 hasFunder F4320338281 @default.
• W2954400107 hasIssue "12" @default.
• W2954400107 hasLocation W29544001071 @default.
• W2954400107 hasOpenAccess W2954400107 @default.
• W2954400107 hasPrimaryLocation W29544001071 @default.
• W2954400107 hasRelatedWork W2961085424 @default.
• W2954400107 hasRelatedWork W3046775127 @default.
• W2954400107 hasRelatedWork W3107602296 @default.
• W2954400107 hasRelatedWork W3170094116 @default.
• W2954400107 hasRelatedWork W3209574120 @default.
• W2954400107 hasRelatedWork W4205958290 @default.
• W2954400107 hasRelatedWork W4286629047 @default.
• W2954400107 hasRelatedWork W4306321456 @default.

• next