Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950678851> ?p ?o ?g. }
- W2950678851 abstract "We discuss a general method to learn data representations from multiple tasks. We provide a justification for this method in both settings of multitask learning and learning-to-learn. The method is illustrated in detail in the special case of linear feature learning. Conditions on the theoretical advantage offered by multitask representation learning over independent task learning are established. In particular, focusing on the important example of half-space learning, we derive the regime in which multitask representation learning is beneficial over independent task learning, as a function of the sample size, the number of tasks and the intrinsic data dimensionality. Other potential applications of our results include multitask feature learning in reproducing kernel Hilbert spaces and multilayer, deep networks." @default.
- W2950678851 created "2019-06-27" @default.
- W2950678851 creator A5008737035 @default.
- W2950678851 creator A5034260726 @default.
- W2950678851 creator A5088601308 @default.
- W2950678851 date "2015-05-23" @default.
- W2950678851 modified "2023-09-27" @default.
- W2950678851 title "The Benefit of Multitask Representation Learning" @default.
- W2950678851 cites W1519789693 @default.
- W2950678851 cites W1541700983 @default.
- W2950678851 cites W1568364756 @default.
- W2950678851 cites W1576642935 @default.
- W2950678851 cites W1608733719 @default.
- W2950678851 cites W1981657694 @default.
- W2950678851 cites W2045313701 @default.
- W2950678851 cites W2047772840 @default.
- W2950678851 cites W2069754508 @default.
- W2950678851 cites W2102605133 @default.
- W2950678851 cites W2103012681 @default.
- W2950678851 cites W2121358800 @default.
- W2950678851 cites W2130903752 @default.
- W2950678851 cites W2154904274 @default.
- W2950678851 cites W2156267734 @default.
- W2950678851 cites W2160943791 @default.
- W2950678851 cites W2165644552 @default.
- W2950678851 cites W2170912114 @default.
- W2950678851 cites W2199934397 @default.
- W2950678851 cites W2237537322 @default.
- W2950678851 cites W2798707604 @default.
- W2950678851 cites W2963917643 @default.
- W2950678851 cites W99485931 @default.
- W2950678851 hasPublicationYear "2015" @default.
- W2950678851 type Work @default.
- W2950678851 sameAs 2950678851 @default.
- W2950678851 citedByCount "3" @default.
- W2950678851 countsByYear W29506788512016 @default.
- W2950678851 crossrefType "posted-content" @default.
- W2950678851 hasAuthorship W2950678851A5008737035 @default.
- W2950678851 hasAuthorship W2950678851A5034260726 @default.
- W2950678851 hasAuthorship W2950678851A5088601308 @default.
- W2950678851 hasConcept C111030470 @default.
- W2950678851 hasConcept C115903097 @default.
- W2950678851 hasConcept C119857082 @default.
- W2950678851 hasConcept C134306372 @default.
- W2950678851 hasConcept C138885662 @default.
- W2950678851 hasConcept C14036430 @default.
- W2950678851 hasConcept C154945302 @default.
- W2950678851 hasConcept C162324750 @default.
- W2950678851 hasConcept C17744445 @default.
- W2950678851 hasConcept C187736073 @default.
- W2950678851 hasConcept C199539241 @default.
- W2950678851 hasConcept C24138899 @default.
- W2950678851 hasConcept C2776359362 @default.
- W2950678851 hasConcept C2776401178 @default.
- W2950678851 hasConcept C2780451532 @default.
- W2950678851 hasConcept C28006648 @default.
- W2950678851 hasConcept C33923547 @default.
- W2950678851 hasConcept C41008148 @default.
- W2950678851 hasConcept C41895202 @default.
- W2950678851 hasConcept C58973888 @default.
- W2950678851 hasConcept C59404180 @default.
- W2950678851 hasConcept C62799726 @default.
- W2950678851 hasConcept C78458016 @default.
- W2950678851 hasConcept C80884492 @default.
- W2950678851 hasConcept C86803240 @default.
- W2950678851 hasConcept C94625758 @default.
- W2950678851 hasConceptScore W2950678851C111030470 @default.
- W2950678851 hasConceptScore W2950678851C115903097 @default.
- W2950678851 hasConceptScore W2950678851C119857082 @default.
- W2950678851 hasConceptScore W2950678851C134306372 @default.
- W2950678851 hasConceptScore W2950678851C138885662 @default.
- W2950678851 hasConceptScore W2950678851C14036430 @default.
- W2950678851 hasConceptScore W2950678851C154945302 @default.
- W2950678851 hasConceptScore W2950678851C162324750 @default.
- W2950678851 hasConceptScore W2950678851C17744445 @default.
- W2950678851 hasConceptScore W2950678851C187736073 @default.
- W2950678851 hasConceptScore W2950678851C199539241 @default.
- W2950678851 hasConceptScore W2950678851C24138899 @default.
- W2950678851 hasConceptScore W2950678851C2776359362 @default.
- W2950678851 hasConceptScore W2950678851C2776401178 @default.
- W2950678851 hasConceptScore W2950678851C2780451532 @default.
- W2950678851 hasConceptScore W2950678851C28006648 @default.
- W2950678851 hasConceptScore W2950678851C33923547 @default.
- W2950678851 hasConceptScore W2950678851C41008148 @default.
- W2950678851 hasConceptScore W2950678851C41895202 @default.
- W2950678851 hasConceptScore W2950678851C58973888 @default.
- W2950678851 hasConceptScore W2950678851C59404180 @default.
- W2950678851 hasConceptScore W2950678851C62799726 @default.
- W2950678851 hasConceptScore W2950678851C78458016 @default.
- W2950678851 hasConceptScore W2950678851C80884492 @default.
- W2950678851 hasConceptScore W2950678851C86803240 @default.
- W2950678851 hasConceptScore W2950678851C94625758 @default.
- W2950678851 hasLocation W29506788511 @default.
- W2950678851 hasOpenAccess W2950678851 @default.
- W2950678851 hasPrimaryLocation W29506788511 @default.
- W2950678851 hasRelatedWork W2132150417 @default.
- W2950678851 hasRelatedWork W2165343651 @default.
- W2950678851 hasRelatedWork W2186862102 @default.
- W2950678851 hasRelatedWork W2407277018 @default.
- W2950678851 hasRelatedWork W251250361 @default.