Matches in SemOpenAlex for { <https://semopenalex.org/work/W3187685774> ?p ?o ?g. }
- W3187685774 endingPage "8239" @default.
- W3187685774 startingPage "8231" @default.
- W3187685774 abstract "The similarity of feature representations plays a pivotal role in the success of problems related to domain adaptation. Feature similarity includes both the invariance of marginal distributions and the closeness of conditional distributions given the desired response y (e.g., class labels). Unfortunately, traditional methods always learn such features without fully taking into consideration the information in y, which in turn may lead to a mismatch of the conditional distributions or the mixup of discriminative structures underlying data distributions. In this work, we introduce the recently proposed von Neumann conditional divergence to improve the transferability across multiple domains. We show that this new divergence is differentiable and eligible to easily quantify the functional dependence between features and y. Given multiple source tasks, we integrate this divergence to capture discriminative information in y and design novel learning objectives assuming those source tasks are observed either simultaneously or sequentially. In both scenarios, we obtain favorable performance against state-of-the-art methods in terms of smaller generalization error on new tasks and less catastrophic forgetting on source tasks (in the sequential setup)." @default.
- W3187685774 created "2021-08-16" @default.
- W3187685774 creator A5030040634 @default.
- W3187685774 creator A5041358503 @default.
- W3187685774 date "2022-06-28" @default.
- W3187685774 modified "2023-09-25" @default.
- W3187685774 title "Learning to Transfer with von Neumann Conditional Divergence" @default.
- W3187685774 cites W144691381 @default.
- W3187685774 cites W1556219185 @default.
- W3187685774 cites W1594031697 @default.
- W3187685774 cites W1604467579 @default.
- W3187685774 cites W1617650991 @default.
- W3187685774 cites W1631356911 @default.
- W3187685774 cites W1965021696 @default.
- W3187685774 cites W1978380814 @default.
- W3187685774 cites W2023320048 @default.
- W3187685774 cites W2097939965 @default.
- W3187685774 cites W2099111195 @default.
- W3187685774 cites W2102201073 @default.
- W3187685774 cites W2104094955 @default.
- W3187685774 cites W2110091014 @default.
- W3187685774 cites W2115403315 @default.
- W3187685774 cites W2120879165 @default.
- W3187685774 cites W2121133842 @default.
- W3187685774 cites W2136922672 @default.
- W3187685774 cites W2149933564 @default.
- W3187685774 cites W2155541015 @default.
- W3187685774 cites W2157444450 @default.
- W3187685774 cites W2163302275 @default.
- W3187685774 cites W2401823607 @default.
- W3187685774 cites W2557579533 @default.
- W3187685774 cites W2560647685 @default.
- W3187685774 cites W2592004655 @default.
- W3187685774 cites W2593597837 @default.
- W3187685774 cites W2683470288 @default.
- W3187685774 cites W2737492962 @default.
- W3187685774 cites W2750384547 @default.
- W3187685774 cites W2786446225 @default.
- W3187685774 cites W2788388592 @default.
- W3187685774 cites W2788768841 @default.
- W3187685774 cites W2880214242 @default.
- W3187685774 cites W2890487450 @default.
- W3187685774 cites W2891975605 @default.
- W3187685774 cites W2892341857 @default.
- W3187685774 cites W2902625698 @default.
- W3187685774 cites W2912280829 @default.
- W3187685774 cites W2913979465 @default.
- W3187685774 cites W2951357534 @default.
- W3187685774 cites W2962724315 @default.
- W3187685774 cites W2963217615 @default.
- W3187685774 cites W2963341924 @default.
- W3187685774 cites W2963390791 @default.
- W3187685774 cites W2963788399 @default.
- W3187685774 cites W2970971581 @default.
- W3187685774 cites W2973152224 @default.
- W3187685774 cites W2986381065 @default.
- W3187685774 cites W2989653627 @default.
- W3187685774 cites W3003536138 @default.
- W3187685774 cites W3006729188 @default.
- W3187685774 cites W3020943070 @default.
- W3187685774 cites W3030364939 @default.
- W3187685774 cites W3034216758 @default.
- W3187685774 cites W3035153354 @default.
- W3187685774 cites W3043058418 @default.
- W3187685774 cites W3097816393 @default.
- W3187685774 cites W3103937687 @default.
- W3187685774 cites W3119697803 @default.
- W3187685774 cites W3120740533 @default.
- W3187685774 cites W3122207284 @default.
- W3187685774 cites W597395834 @default.
- W3187685774 doi "https://doi.org/10.1609/aaai.v36i8.20797" @default.
- W3187685774 hasPublicationYear "2022" @default.
- W3187685774 type Work @default.
- W3187685774 sameAs 3187685774 @default.
- W3187685774 citedByCount "0" @default.
- W3187685774 crossrefType "journal-article" @default.
- W3187685774 hasAuthorship W3187685774A5030040634 @default.
- W3187685774 hasAuthorship W3187685774A5041358503 @default.
- W3187685774 hasBestOaLocation W31876857741 @default.
- W3187685774 hasConcept C103278499 @default.
- W3187685774 hasConcept C105795698 @default.
- W3187685774 hasConcept C115961682 @default.
- W3187685774 hasConcept C119857082 @default.
- W3187685774 hasConcept C122123141 @default.
- W3187685774 hasConcept C134306372 @default.
- W3187685774 hasConcept C138885662 @default.
- W3187685774 hasConcept C153180895 @default.
- W3187685774 hasConcept C154945302 @default.
- W3187685774 hasConcept C165216359 @default.
- W3187685774 hasConcept C177148314 @default.
- W3187685774 hasConcept C202615002 @default.
- W3187685774 hasConcept C207390915 @default.
- W3187685774 hasConcept C2776401178 @default.
- W3187685774 hasConcept C2779545769 @default.
- W3187685774 hasConcept C33923547 @default.
- W3187685774 hasConcept C41008148 @default.
- W3187685774 hasConcept C41895202 @default.
- W3187685774 hasConcept C43555835 @default.