Matches in SemOpenAlex for { <https://semopenalex.org/work/W2964055354> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W2964055354 endingPage "436" @default.
- W2964055354 startingPage "420" @default.
- W2964055354 abstract "In this paper, we make two contributions to unsupervised domain adaptation (UDA) using the convolutional neural network (CNN). First, our approach transfers knowledge in all the convolutional layers through attention alignment. Most previous methods align high-level representations, e.g., activations of the fully connected (FC) layers. In these methods, however, the convolutional layers which underpin critical low-level domain knowledge cannot be updated directly towards reducing domain discrepancy. Specifically, we assume that the discriminative regions in an image are relatively invariant to image style changes. Based on this assumption, we propose an attention alignment scheme on all the target convolutional layers to uncover the knowledge shared by the source domain. Second, we estimate the posterior label distribution of the unlabeled data for target network training. Previous methods, which iteratively update the pseudo labels by the target network and refine the target network by the updated pseudo labels, are vulnerable to label estimation errors. Instead, our approach uses category distribution to calculate the cross-entropy loss for training, thereby ameliorating the error accumulation of the estimated labels. The two contributions allow our approach to outperform the state-of-the-art methods by +2.6% on the Office-31 dataset." @default.
- W2964055354 created "2019-07-30" @default.
- W2964055354 creator A5005421447 @default.
- W2964055354 creator A5011488839 @default.
- W2964055354 creator A5048411184 @default.
- W2964055354 creator A5053773821 @default.
- W2964055354 date "2018-01-01" @default.
- W2964055354 modified "2023-10-18" @default.
- W2964055354 title "Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: The Benefit of Target Expectation Maximization" @default.
- W2964055354 cites W1849277567 @default.
- W2964055354 cites W2088801701 @default.
- W2964055354 cites W2108598243 @default.
- W2964055354 cites W2110158442 @default.
- W2964055354 cites W2112796928 @default.
- W2964055354 cites W2194775991 @default.
- W2964055354 cites W2214409633 @default.
- W2964055354 cites W2295107390 @default.
- W2964055354 cites W2302255633 @default.
- W2964055354 cites W2584009249 @default.
- W2964055354 cites W2585635281 @default.
- W2964055354 cites W2593768305 @default.
- W2964055354 cites W2600144439 @default.
- W2964055354 cites W2617027347 @default.
- W2964055354 cites W2744915377 @default.
- W2964055354 cites W2962793481 @default.
- W2964055354 cites W2962858109 @default.
- W2964055354 cites W2962925415 @default.
- W2964055354 cites W2963709863 @default.
- W2964055354 cites W2964274719 @default.
- W2964055354 cites W2964288524 @default.
- W2964055354 doi "https://doi.org/10.1007/978-3-030-01252-6_25" @default.
- W2964055354 hasPublicationYear "2018" @default.
- W2964055354 type Work @default.
- W2964055354 sameAs 2964055354 @default.
- W2964055354 citedByCount "93" @default.
- W2964055354 countsByYear W29640553542018 @default.
- W2964055354 countsByYear W29640553542019 @default.
- W2964055354 countsByYear W29640553542020 @default.
- W2964055354 countsByYear W29640553542021 @default.
- W2964055354 countsByYear W29640553542022 @default.
- W2964055354 countsByYear W29640553542023 @default.
- W2964055354 crossrefType "book-chapter" @default.
- W2964055354 hasAuthorship W2964055354A5005421447 @default.
- W2964055354 hasAuthorship W2964055354A5011488839 @default.
- W2964055354 hasAuthorship W2964055354A5048411184 @default.
- W2964055354 hasAuthorship W2964055354A5053773821 @default.
- W2964055354 hasBestOaLocation W29640553542 @default.
- W2964055354 hasConcept C126255220 @default.
- W2964055354 hasConcept C134306372 @default.
- W2964055354 hasConcept C139807058 @default.
- W2964055354 hasConcept C154945302 @default.
- W2964055354 hasConcept C15744967 @default.
- W2964055354 hasConcept C169760540 @default.
- W2964055354 hasConcept C2776330181 @default.
- W2964055354 hasConcept C2776434776 @default.
- W2964055354 hasConcept C33923547 @default.
- W2964055354 hasConcept C36503486 @default.
- W2964055354 hasConcept C37736160 @default.
- W2964055354 hasConcept C41008148 @default.
- W2964055354 hasConcept C95623464 @default.
- W2964055354 hasConceptScore W2964055354C126255220 @default.
- W2964055354 hasConceptScore W2964055354C134306372 @default.
- W2964055354 hasConceptScore W2964055354C139807058 @default.
- W2964055354 hasConceptScore W2964055354C154945302 @default.
- W2964055354 hasConceptScore W2964055354C15744967 @default.
- W2964055354 hasConceptScore W2964055354C169760540 @default.
- W2964055354 hasConceptScore W2964055354C2776330181 @default.
- W2964055354 hasConceptScore W2964055354C2776434776 @default.
- W2964055354 hasConceptScore W2964055354C33923547 @default.
- W2964055354 hasConceptScore W2964055354C36503486 @default.
- W2964055354 hasConceptScore W2964055354C37736160 @default.
- W2964055354 hasConceptScore W2964055354C41008148 @default.
- W2964055354 hasConceptScore W2964055354C95623464 @default.
- W2964055354 hasLocation W29640553541 @default.
- W2964055354 hasLocation W29640553542 @default.
- W2964055354 hasLocation W29640553543 @default.
- W2964055354 hasOpenAccess W2964055354 @default.
- W2964055354 hasPrimaryLocation W29640553541 @default.
- W2964055354 hasRelatedWork W2152148513 @default.
- W2964055354 hasRelatedWork W2892202790 @default.
- W2964055354 hasRelatedWork W2913957957 @default.
- W2964055354 hasRelatedWork W2951876757 @default.
- W2964055354 hasRelatedWork W2955455867 @default.
- W2964055354 hasRelatedWork W3017503936 @default.
- W2964055354 hasRelatedWork W3037757213 @default.
- W2964055354 hasRelatedWork W3100356292 @default.
- W2964055354 hasRelatedWork W3174713482 @default.
- W2964055354 hasRelatedWork W4281397339 @default.
- W2964055354 isParatext "false" @default.
- W2964055354 isRetracted "false" @default.
- W2964055354 magId "2964055354" @default.
- W2964055354 workType "book-chapter" @default.