FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2936103337> ?p ?o ?g. }

• W2936103337 abstract "Several tasks in machine learning are evaluated using non-differentiable metrics such as mean average precision or Spearman correlation. However, their non-differentiability prevents from using them as objective functions in a learning framework. Surrogate and relaxation methods exist but tend to be specific to a given metric. In the present work, we introduce a new method to learn approximations of such non-differentiable objective functions. Our approach is based on a deep architecture that approximates the sorting of arbitrary sets of scores. It is trained virtually for free using synthetic data. This sorting deep (SoDeep) net can then be combined in a plug-and-play manner with existing deep architectures. We demonstrate the interest of our approach in three different tasks that require ranking: Cross-modal text-image retrieval, multi-label image classification and visual memorability ranking. Our approach yields very competitive results on these three tasks, which validates the merit and the flexibility of SoDeep as a proxy for sorting operation in ranking-based losses." @default.
• W2936103337 created "2019-04-25" @default.
• W2936103337 creator A5022871131 @default.
• W2936103337 creator A5027039225 @default.
• W2936103337 creator A5052035887 @default.
• W2936103337 creator A5076170578 @default.
• W2936103337 date "2019-04-08" @default.
• W2936103337 modified "2023-09-25" @default.
• W2936103337 title "SoDeep: a Sorting Deep net to learn ranking loss surrogates." @default.
• W2936103337 cites W1504663295 @default.
• W2936103337 cites W1527575280 @default.
• W2936103337 cites W1532499126 @default.
• W2936103337 cites W1686810756 @default.
• W2936103337 cites W1861492603 @default.
• W2936103337 cites W1905882502 @default.
• W2936103337 cites W1916445035 @default.
• W2936103337 cites W1944672 @default.
• W2936103337 cites W2005063569 @default.
• W2936103337 cites W2040870580 @default.
• W2936103337 cites W2047221353 @default.
• W2936103337 cites W2083905053 @default.
• W2936103337 cites W2106053110 @default.
• W2936103337 cites W2108862644 @default.
• W2936103337 cites W2117154949 @default.
• W2936103337 cites W2123024445 @default.
• W2936103337 cites W2124386111 @default.
• W2936103337 cites W2127176025 @default.
• W2936103337 cites W2131774270 @default.
• W2936103337 cites W2140634465 @default.
• W2936103337 cites W2143331230 @default.
• W2936103337 cites W2163605009 @default.
• W2936103337 cites W2194775991 @default.
• W2936103337 cites W2208178861 @default.
• W2936103337 cites W2219771564 @default.
• W2936103337 cites W2478090196 @default.
• W2936103337 cites W2606473278 @default.
• W2936103337 cites W2738853914 @default.
• W2936103337 cites W2739181657 @default.
• W2936103337 cites W2751185861 @default.
• W2936103337 cites W2811504762 @default.
• W2936103337 cites W2907044893 @default.
• W2936103337 cites W2908440169 @default.
• W2936103337 cites W2949117887 @default.
• W2936103337 cites W2963467339 @default.
• W2936103337 cites W2963789586 @default.
• W2936103337 cites W2964121744 @default.
• W2936103337 hasPublicationYear "2019" @default.
• W2936103337 type Work @default.
• W2936103337 sameAs 2936103337 @default.
• W2936103337 citedByCount "2" @default.
• W2936103337 countsByYear W29361033372020 @default.
• W2936103337 countsByYear W29361033372021 @default.
• W2936103337 crossrefType "posted-content" @default.
• W2936103337 hasAuthorship W2936103337A5022871131 @default.
• W2936103337 hasAuthorship W2936103337A5027039225 @default.
• W2936103337 hasAuthorship W2936103337A5052035887 @default.
• W2936103337 hasAuthorship W2936103337A5076170578 @default.
• W2936103337 hasConcept C108583219 @default.
• W2936103337 hasConcept C111696304 @default.
• W2936103337 hasConcept C11413529 @default.
• W2936103337 hasConcept C119857082 @default.
• W2936103337 hasConcept C124101348 @default.
• W2936103337 hasConcept C134306372 @default.
• W2936103337 hasConcept C153180895 @default.
• W2936103337 hasConcept C154945302 @default.
• W2936103337 hasConcept C162324750 @default.
• W2936103337 hasConcept C176217482 @default.
• W2936103337 hasConcept C189430467 @default.
• W2936103337 hasConcept C202615002 @default.
• W2936103337 hasConcept C21547014 @default.
• W2936103337 hasConcept C33923547 @default.
• W2936103337 hasConcept C41008148 @default.
• W2936103337 hasConceptScore W2936103337C108583219 @default.
• W2936103337 hasConceptScore W2936103337C111696304 @default.
• W2936103337 hasConceptScore W2936103337C11413529 @default.
• W2936103337 hasConceptScore W2936103337C119857082 @default.
• W2936103337 hasConceptScore W2936103337C124101348 @default.
• W2936103337 hasConceptScore W2936103337C134306372 @default.
• W2936103337 hasConceptScore W2936103337C153180895 @default.
• W2936103337 hasConceptScore W2936103337C154945302 @default.
• W2936103337 hasConceptScore W2936103337C162324750 @default.
• W2936103337 hasConceptScore W2936103337C176217482 @default.
• W2936103337 hasConceptScore W2936103337C189430467 @default.
• W2936103337 hasConceptScore W2936103337C202615002 @default.
• W2936103337 hasConceptScore W2936103337C21547014 @default.
• W2936103337 hasConceptScore W2936103337C33923547 @default.
• W2936103337 hasConceptScore W2936103337C41008148 @default.
• W2936103337 hasOpenAccess W2936103337 @default.
• W2936103337 hasRelatedWork W1959567017 @default.
• W2936103337 hasRelatedWork W2022177107 @default.
• W2936103337 hasRelatedWork W2045260043 @default.
• W2936103337 hasRelatedWork W2129735430 @default.
• W2936103337 hasRelatedWork W2167257437 @default.
• W2936103337 hasRelatedWork W2604167613 @default.
• W2936103337 hasRelatedWork W2604829508 @default.
• W2936103337 hasRelatedWork W2764149444 @default.
• W2936103337 hasRelatedWork W2798785178 @default.
• W2936103337 hasRelatedWork W2876724543 @default.
• W2936103337 hasRelatedWork W2892611525 @default.
• W2936103337 hasRelatedWork W2896348918 @default.

• next