FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2075536623> ?p ?o ?g. }

• W2075536623 abstract "We describe a novel max-margin parameter learning approach for structured prediction problems under certain non-decomposable performance measures. Structured prediction is a common approach in many vision problems. Non-decomposable performance measures are also commonplace. However, efficient general methods for learning parameters against non-decomposable performance measures do not exist. In this paper we develop such a method, based on dual decomposition, that is applicable to a large class of non-decomposable performance measures. We exploit dual decomposition to factorize the original hard problem into two smaller problems and show how to optimize each factor efficiently. We show experimentally that the proposed approach significantly outperforms alternatives, which either sacrifice the model structure or approximate the performance measure, and is an order of magnitude faster than a previous approach with comparable results." @default.
• W2075536623 created "2016-06-24" @default.
• W2075536623 creator A5017638706 @default.
• W2075536623 creator A5038984764 @default.
• W2075536623 creator A5080112958 @default.
• W2075536623 date "2012-06-01" @default.
• W2075536623 modified "2023-10-18" @default.
• W2075536623 title "Complex loss optimization via dual decomposition" @default.
• W2075536623 cites W1518470882 @default.
• W2075536623 cites W1970549169 @default.
• W2075536623 cites W1978511849 @default.
• W2075536623 cites W1999478155 @default.
• W2075536623 cites W2031489346 @default.
• W2075536623 cites W2070771761 @default.
• W2075536623 cites W2097826433 @default.
• W2075536623 cites W2127176025 @default.
• W2075536623 cites W2131538180 @default.
• W2075536623 cites W2135414191 @default.
• W2075536623 cites W2137880010 @default.
• W2075536623 cites W2140634465 @default.
• W2075536623 cites W2142037471 @default.
• W2075536623 cites W2142623206 @default.
• W2075536623 cites W2143516773 @default.
• W2075536623 cites W2162670331 @default.
• W2075536623 cites W2162762921 @default.
• W2075536623 cites W2168356304 @default.
• W2075536623 cites W2429914308 @default.
• W2075536623 doi "https://doi.org/10.1109/cvpr.2012.6247941" @default.
• W2075536623 hasPublicationYear "2012" @default.
• W2075536623 type Work @default.
• W2075536623 sameAs 2075536623 @default.
• W2075536623 citedByCount "8" @default.
• W2075536623 countsByYear W20755366232013 @default.
• W2075536623 countsByYear W20755366232015 @default.
• W2075536623 countsByYear W20755366232017 @default.
• W2075536623 countsByYear W20755366232019 @default.
• W2075536623 countsByYear W20755366232023 @default.
• W2075536623 crossrefType "proceedings-article" @default.
• W2075536623 hasAuthorship W2075536623A5017638706 @default.
• W2075536623 hasAuthorship W2075536623A5038984764 @default.
• W2075536623 hasAuthorship W2075536623A5080112958 @default.
• W2075536623 hasBestOaLocation W20755366232 @default.
• W2075536623 hasConcept C11413529 @default.
• W2075536623 hasConcept C119857082 @default.
• W2075536623 hasConcept C121332964 @default.
• W2075536623 hasConcept C124101348 @default.
• W2075536623 hasConcept C124681953 @default.
• W2075536623 hasConcept C124952713 @default.
• W2075536623 hasConcept C126255220 @default.
• W2075536623 hasConcept C137836250 @default.
• W2075536623 hasConcept C142362112 @default.
• W2075536623 hasConcept C154945302 @default.
• W2075536623 hasConcept C158693339 @default.
• W2075536623 hasConcept C165696696 @default.
• W2075536623 hasConcept C187834632 @default.
• W2075536623 hasConcept C18903297 @default.
• W2075536623 hasConcept C2777212361 @default.
• W2075536623 hasConcept C2780009758 @default.
• W2075536623 hasConcept C2780980858 @default.
• W2075536623 hasConcept C33923547 @default.
• W2075536623 hasConcept C38652104 @default.
• W2075536623 hasConcept C41008148 @default.
• W2075536623 hasConcept C42355184 @default.
• W2075536623 hasConcept C62520636 @default.
• W2075536623 hasConcept C774472 @default.
• W2075536623 hasConcept C86803240 @default.
• W2075536623 hasConceptScore W2075536623C11413529 @default.
• W2075536623 hasConceptScore W2075536623C119857082 @default.
• W2075536623 hasConceptScore W2075536623C121332964 @default.
• W2075536623 hasConceptScore W2075536623C124101348 @default.
• W2075536623 hasConceptScore W2075536623C124681953 @default.
• W2075536623 hasConceptScore W2075536623C124952713 @default.
• W2075536623 hasConceptScore W2075536623C126255220 @default.
• W2075536623 hasConceptScore W2075536623C137836250 @default.
• W2075536623 hasConceptScore W2075536623C142362112 @default.
• W2075536623 hasConceptScore W2075536623C154945302 @default.
• W2075536623 hasConceptScore W2075536623C158693339 @default.
• W2075536623 hasConceptScore W2075536623C165696696 @default.
• W2075536623 hasConceptScore W2075536623C187834632 @default.
• W2075536623 hasConceptScore W2075536623C18903297 @default.
• W2075536623 hasConceptScore W2075536623C2777212361 @default.
• W2075536623 hasConceptScore W2075536623C2780009758 @default.
• W2075536623 hasConceptScore W2075536623C2780980858 @default.
• W2075536623 hasConceptScore W2075536623C33923547 @default.
• W2075536623 hasConceptScore W2075536623C38652104 @default.
• W2075536623 hasConceptScore W2075536623C41008148 @default.
• W2075536623 hasConceptScore W2075536623C42355184 @default.
• W2075536623 hasConceptScore W2075536623C62520636 @default.
• W2075536623 hasConceptScore W2075536623C774472 @default.
• W2075536623 hasConceptScore W2075536623C86803240 @default.
• W2075536623 hasLocation W20755366231 @default.
• W2075536623 hasLocation W20755366232 @default.
• W2075536623 hasOpenAccess W2075536623 @default.
• W2075536623 hasPrimaryLocation W20755366231 @default.
• W2075536623 hasRelatedWork W1988437637 @default.
• W2075536623 hasRelatedWork W2075536623 @default.
• W2075536623 hasRelatedWork W2176073131 @default.
• W2075536623 hasRelatedWork W2275167993 @default.
• W2075536623 hasRelatedWork W2897430141 @default.
• W2075536623 hasRelatedWork W2962781798 @default.

• next