FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2116418767> ?p ?o ?g. }

• W2116418767 endingPage "1313" @default.
• W2116418767 startingPage "1305" @default.
• W2116418767 abstract "Recently it has been shown that classical supervised and unsupervised training methods can be unified as special cases of so-called “optimal reverse prediction”: predicting inputs from target labels while optimizing over both model parameters and missing labels. Although this perspective establishes links between classical training principles, the existing formulation only applies to linear predictors under squared loss, hence is extremely limited. We generalize the formulation of optimal reverse prediction to arbitrary Bregman divergences, and more importantly to non-linear predictors. This extension is achieved by establishing a generalized form of forwardreverse minimization equivalence that holds for arbitrary matching losses. Several benefits follow. First, a new variant of Bregman divergence clustering can be recovered that incorporates a non-linear data reconstruction model. Second, normalized-cut and kernel-based extensions can be formulated coherently. Finally, a new semi-supervised training principle can be recovered for classification problems that demonstrates some advantages over the state of the art." @default.
• W2116418767 created "2016-06-24" @default.
• W2116418767 creator A5010575626 @default.
• W2116418767 creator A5031606152 @default.
• W2116418767 date "2012-03-21" @default.
• W2116418767 modified "2023-09-26" @default.
• W2116418767 title "Generalized Optimal Reverse Prediction" @default.
• W2116418767 cites W1529879082 @default.
• W2116418767 cites W1540198634 @default.
• W2116418767 cites W1578488390 @default.
• W2116418767 cites W1589919686 @default.
• W2116418767 cites W1663973292 @default.
• W2116418767 cites W1822246767 @default.
• W2116418767 cites W2015904350 @default.
• W2116418767 cites W2028265899 @default.
• W2116418767 cites W2048679005 @default.
• W2116418767 cites W2058115154 @default.
• W2116418767 cites W2069663974 @default.
• W2116418767 cites W2096765209 @default.
• W2116418767 cites W2097089247 @default.
• W2116418767 cites W2101762657 @default.
• W2116418767 cites W2104290444 @default.
• W2116418767 cites W2104755048 @default.
• W2116418767 cites W2106579555 @default.
• W2116418767 cites W2107008379 @default.
• W2116418767 cites W2109501666 @default.
• W2116418767 cites W2111557120 @default.
• W2116418767 cites W2113484479 @default.
• W2116418767 cites W2135001774 @default.
• W2116418767 cites W2136504847 @default.
• W2116418767 cites W2141996170 @default.
• W2116418767 cites W2148029428 @default.
• W2116418767 cites W2153939756 @default.
• W2116418767 cites W2154455818 @default.
• W2116418767 cites W2163046677 @default.
• W2116418767 cites W2170938446 @default.
• W2116418767 cites W2183800414 @default.
• W2116418767 cites W2252194958 @default.
• W2116418767 cites W2501155501 @default.
• W2116418767 cites W2610360795 @default.
• W2116418767 hasPublicationYear "2012" @default.
• W2116418767 type Work @default.
• W2116418767 sameAs 2116418767 @default.
• W2116418767 citedByCount "5" @default.
• W2116418767 countsByYear W21164187672012 @default.
• W2116418767 countsByYear W21164187672014 @default.
• W2116418767 countsByYear W21164187672015 @default.
• W2116418767 countsByYear W21164187672016 @default.
• W2116418767 countsByYear W21164187672019 @default.
• W2116418767 crossrefType "proceedings-article" @default.
• W2116418767 hasAuthorship W2116418767A5010575626 @default.
• W2116418767 hasAuthorship W2116418767A5031606152 @default.
• W2116418767 hasConcept C11413529 @default.
• W2116418767 hasConcept C114614502 @default.
• W2116418767 hasConcept C118615104 @default.
• W2116418767 hasConcept C119857082 @default.
• W2116418767 hasConcept C126255220 @default.
• W2116418767 hasConcept C138885662 @default.
• W2116418767 hasConcept C147764199 @default.
• W2116418767 hasConcept C149073432 @default.
• W2116418767 hasConcept C154945302 @default.
• W2116418767 hasConcept C199360897 @default.
• W2116418767 hasConcept C207390915 @default.
• W2116418767 hasConcept C2778029271 @default.
• W2116418767 hasConcept C2780069185 @default.
• W2116418767 hasConcept C28826006 @default.
• W2116418767 hasConcept C33923547 @default.
• W2116418767 hasConcept C41008148 @default.
• W2116418767 hasConcept C41895202 @default.
• W2116418767 hasConcept C73555534 @default.
• W2116418767 hasConcept C74193536 @default.
• W2116418767 hasConceptScore W2116418767C11413529 @default.
• W2116418767 hasConceptScore W2116418767C114614502 @default.
• W2116418767 hasConceptScore W2116418767C118615104 @default.
• W2116418767 hasConceptScore W2116418767C119857082 @default.
• W2116418767 hasConceptScore W2116418767C126255220 @default.
• W2116418767 hasConceptScore W2116418767C138885662 @default.
• W2116418767 hasConceptScore W2116418767C147764199 @default.
• W2116418767 hasConceptScore W2116418767C149073432 @default.
• W2116418767 hasConceptScore W2116418767C154945302 @default.
• W2116418767 hasConceptScore W2116418767C199360897 @default.
• W2116418767 hasConceptScore W2116418767C207390915 @default.
• W2116418767 hasConceptScore W2116418767C2778029271 @default.
• W2116418767 hasConceptScore W2116418767C2780069185 @default.
• W2116418767 hasConceptScore W2116418767C28826006 @default.
• W2116418767 hasConceptScore W2116418767C33923547 @default.
• W2116418767 hasConceptScore W2116418767C41008148 @default.
• W2116418767 hasConceptScore W2116418767C41895202 @default.
• W2116418767 hasConceptScore W2116418767C73555534 @default.
• W2116418767 hasConceptScore W2116418767C74193536 @default.
• W2116418767 hasLocation W21164187671 @default.
• W2116418767 hasOpenAccess W2116418767 @default.
• W2116418767 hasPrimaryLocation W21164187671 @default.
• W2116418767 hasRelatedWork W2082319640 @default.
• W2116418767 hasRelatedWork W2133095989 @default.
• W2116418767 hasRelatedWork W2147121386 @default.
• W2116418767 hasRelatedWork W2169477532 @default.
• W2116418767 hasRelatedWork W2250542744 @default.

• next