Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312786518> ?p ?o ?g. }
Showing items 1 to 96 of
96
with 100 items per page.
- W4312786518 abstract "Deep learning methods can struggle to handle domain shifts not seen in training data, which can cause them to not generalize well to unseen domains. This has led to research attention on domain generalization (DG), which aims to the model's generalization ability to out-of-distribution. Adversarial domain generalization is a popular approach to DG, but conventional approaches (1) struggle to sufficiently align features so that local neighborhoods are mixed across domains; and (2) can suffer from feature space over collapse which can threaten generalization performance. To address these limitations, we propose localized adversarial domain generalization with space compactness maintenance (LADG) which constitutes two major contributions. First, we propose an adversarial localized classifier as the domain discriminator, along with a principled primary branch. This constructs a min-max game whereby the aim of the featurizer is to produce locally mixed domains. Second, we propose to use a coding-rate loss to alleviate feature space over collapse. We conduct comprehensive experiments on the Wilds DG benchmark to validate our approach, where LADG outperforms leading competitors on most datasets." @default.
- W4312786518 created "2023-01-05" @default.
- W4312786518 creator A5019046203 @default.
- W4312786518 creator A5045227579 @default.
- W4312786518 creator A5057328085 @default.
- W4312786518 creator A5058215578 @default.
- W4312786518 creator A5077894483 @default.
- W4312786518 creator A5082664098 @default.
- W4312786518 date "2022-06-01" @default.
- W4312786518 modified "2023-09-30" @default.
- W4312786518 title "Localized Adversarial Domain Generalization" @default.
- W4312786518 cites W1967004441 @default.
- W4312786518 cites W2069437695 @default.
- W4312786518 cites W2164931791 @default.
- W4312786518 cites W2584886900 @default.
- W4312786518 cites W2763549966 @default.
- W4312786518 cites W2798658180 @default.
- W4312786518 cites W2889232360 @default.
- W4312786518 cites W2911964244 @default.
- W4312786518 cites W2920807444 @default.
- W4312786518 cites W2963043696 @default.
- W4312786518 cites W2963116854 @default.
- W4312786518 cites W2963118547 @default.
- W4312786518 cites W2963350032 @default.
- W4312786518 cites W2963785576 @default.
- W4312786518 cites W2971196067 @default.
- W4312786518 cites W2979805229 @default.
- W4312786518 cites W2998229299 @default.
- W4312786518 cites W2998653622 @default.
- W4312786518 cites W2998712190 @default.
- W4312786518 cites W3016179654 @default.
- W4312786518 cites W3027716283 @default.
- W4312786518 cites W3095253761 @default.
- W4312786518 cites W3153511633 @default.
- W4312786518 cites W3176732771 @default.
- W4312786518 cites W3190208102 @default.
- W4312786518 doi "https://doi.org/10.1109/cvpr52688.2022.00697" @default.
- W4312786518 hasPublicationYear "2022" @default.
- W4312786518 type Work @default.
- W4312786518 citedByCount "3" @default.
- W4312786518 countsByYear W43127865182023 @default.
- W4312786518 crossrefType "proceedings-article" @default.
- W4312786518 hasAuthorship W4312786518A5019046203 @default.
- W4312786518 hasAuthorship W4312786518A5045227579 @default.
- W4312786518 hasAuthorship W4312786518A5057328085 @default.
- W4312786518 hasAuthorship W4312786518A5058215578 @default.
- W4312786518 hasAuthorship W4312786518A5077894483 @default.
- W4312786518 hasAuthorship W4312786518A5082664098 @default.
- W4312786518 hasBestOaLocation W43127865182 @default.
- W4312786518 hasConcept C119857082 @default.
- W4312786518 hasConcept C134306372 @default.
- W4312786518 hasConcept C138885662 @default.
- W4312786518 hasConcept C154945302 @default.
- W4312786518 hasConcept C177148314 @default.
- W4312786518 hasConcept C2776401178 @default.
- W4312786518 hasConcept C2779803651 @default.
- W4312786518 hasConcept C33923547 @default.
- W4312786518 hasConcept C36503486 @default.
- W4312786518 hasConcept C37736160 @default.
- W4312786518 hasConcept C41008148 @default.
- W4312786518 hasConcept C41895202 @default.
- W4312786518 hasConcept C76155785 @default.
- W4312786518 hasConcept C94915269 @default.
- W4312786518 hasConcept C95623464 @default.
- W4312786518 hasConceptScore W4312786518C119857082 @default.
- W4312786518 hasConceptScore W4312786518C134306372 @default.
- W4312786518 hasConceptScore W4312786518C138885662 @default.
- W4312786518 hasConceptScore W4312786518C154945302 @default.
- W4312786518 hasConceptScore W4312786518C177148314 @default.
- W4312786518 hasConceptScore W4312786518C2776401178 @default.
- W4312786518 hasConceptScore W4312786518C2779803651 @default.
- W4312786518 hasConceptScore W4312786518C33923547 @default.
- W4312786518 hasConceptScore W4312786518C36503486 @default.
- W4312786518 hasConceptScore W4312786518C37736160 @default.
- W4312786518 hasConceptScore W4312786518C41008148 @default.
- W4312786518 hasConceptScore W4312786518C41895202 @default.
- W4312786518 hasConceptScore W4312786518C76155785 @default.
- W4312786518 hasConceptScore W4312786518C94915269 @default.
- W4312786518 hasConceptScore W4312786518C95623464 @default.
- W4312786518 hasLocation W43127865181 @default.
- W4312786518 hasLocation W43127865182 @default.
- W4312786518 hasOpenAccess W4312786518 @default.
- W4312786518 hasPrimaryLocation W43127865181 @default.
- W4312786518 hasRelatedWork W2412510955 @default.
- W4312786518 hasRelatedWork W2783726038 @default.
- W4312786518 hasRelatedWork W2803949585 @default.
- W4312786518 hasRelatedWork W2884022884 @default.
- W4312786518 hasRelatedWork W2922272129 @default.
- W4312786518 hasRelatedWork W2924282518 @default.
- W4312786518 hasRelatedWork W2981515171 @default.
- W4312786518 hasRelatedWork W3041381337 @default.
- W4312786518 hasRelatedWork W3196098778 @default.
- W4312786518 hasRelatedWork W3211250490 @default.
- W4312786518 isParatext "false" @default.
- W4312786518 isRetracted "false" @default.
- W4312786518 workType "article" @default.