Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951495199> ?p ?o ?g. }
- W2951495199 abstract "Recently, learning discriminative features to improve the recognition performances gradually becomes the primary goal of deep learning, and numerous remarkable works have emerged. In this paper, we propose a novel yet extremely simple method textbf{Virtual Softmax} to enhance the discriminative property of learned features by injecting a dynamic virtual negative class into the original softmax. Injecting virtual class aims to enlarge inter-class margin and compress intra-class distribution by strengthening the decision boundary constraint. Although it seems weird to optimize with this additional virtual class, we show that our method derives from an intuitive and clear motivation, and it indeed encourages the features to be more compact and separable. This paper empirically and experimentally demonstrates the superiority of Virtual Softmax, improving the performances on a variety of object classification and face verification tasks." @default.
- W2951495199 created "2019-06-27" @default.
- W2951495199 creator A5025452586 @default.
- W2951495199 creator A5058463253 @default.
- W2951495199 creator A5084809225 @default.
- W2951495199 date "2018-11-30" @default.
- W2951495199 modified "2023-09-23" @default.
- W2951495199 title "Virtual Class Enhanced Discriminative Embedding Learning" @default.
- W2951495199 cites W1616462885 @default.
- W2951495199 cites W1782590233 @default.
- W2951495199 cites W1898560071 @default.
- W2951495199 cites W1934184906 @default.
- W2951495199 cites W1950843348 @default.
- W2951495199 cites W2097117768 @default.
- W2951495199 cites W2117539524 @default.
- W2951495199 cites W2183341477 @default.
- W2951495199 cites W2194775991 @default.
- W2951495199 cites W2325939864 @default.
- W2951495199 cites W2335728318 @default.
- W2951495199 cites W2401231614 @default.
- W2951495199 cites W2511730936 @default.
- W2951495199 cites W2520774990 @default.
- W2951495199 cites W2555897561 @default.
- W2951495199 cites W2557864411 @default.
- W2951495199 cites W2562863628 @default.
- W2951495199 cites W2594407953 @default.
- W2951495199 cites W2609296554 @default.
- W2951495199 cites W2741430497 @default.
- W2951495199 cites W2750562598 @default.
- W2951495199 cites W2752782242 @default.
- W2951495199 cites W2950094539 @default.
- W2951495199 cites W2951093852 @default.
- W2951495199 cites W2951143163 @default.
- W2951495199 cites W2952419167 @default.
- W2951495199 cites W2963026686 @default.
- W2951495199 cites W2963606038 @default.
- W2951495199 cites W2963919294 @default.
- W2951495199 cites W3118608800 @default.
- W2951495199 cites W14333344 @default.
- W2951495199 hasPublicationYear "2018" @default.
- W2951495199 type Work @default.
- W2951495199 sameAs 2951495199 @default.
- W2951495199 citedByCount "2" @default.
- W2951495199 countsByYear W29514951992020 @default.
- W2951495199 countsByYear W29514951992021 @default.
- W2951495199 crossrefType "posted-content" @default.
- W2951495199 hasAuthorship W2951495199A5025452586 @default.
- W2951495199 hasAuthorship W2951495199A5058463253 @default.
- W2951495199 hasAuthorship W2951495199A5084809225 @default.
- W2951495199 hasConcept C108583219 @default.
- W2951495199 hasConcept C111472728 @default.
- W2951495199 hasConcept C119857082 @default.
- W2951495199 hasConcept C136197465 @default.
- W2951495199 hasConcept C138885662 @default.
- W2951495199 hasConcept C144024400 @default.
- W2951495199 hasConcept C153180895 @default.
- W2951495199 hasConcept C154945302 @default.
- W2951495199 hasConcept C188441871 @default.
- W2951495199 hasConcept C189950617 @default.
- W2951495199 hasConcept C2524010 @default.
- W2951495199 hasConcept C2776036281 @default.
- W2951495199 hasConcept C2777212361 @default.
- W2951495199 hasConcept C2779304628 @default.
- W2951495199 hasConcept C2780586882 @default.
- W2951495199 hasConcept C2781238097 @default.
- W2951495199 hasConcept C33923547 @default.
- W2951495199 hasConcept C36289849 @default.
- W2951495199 hasConcept C41008148 @default.
- W2951495199 hasConcept C41608201 @default.
- W2951495199 hasConcept C774472 @default.
- W2951495199 hasConcept C97931131 @default.
- W2951495199 hasConceptScore W2951495199C108583219 @default.
- W2951495199 hasConceptScore W2951495199C111472728 @default.
- W2951495199 hasConceptScore W2951495199C119857082 @default.
- W2951495199 hasConceptScore W2951495199C136197465 @default.
- W2951495199 hasConceptScore W2951495199C138885662 @default.
- W2951495199 hasConceptScore W2951495199C144024400 @default.
- W2951495199 hasConceptScore W2951495199C153180895 @default.
- W2951495199 hasConceptScore W2951495199C154945302 @default.
- W2951495199 hasConceptScore W2951495199C188441871 @default.
- W2951495199 hasConceptScore W2951495199C189950617 @default.
- W2951495199 hasConceptScore W2951495199C2524010 @default.
- W2951495199 hasConceptScore W2951495199C2776036281 @default.
- W2951495199 hasConceptScore W2951495199C2777212361 @default.
- W2951495199 hasConceptScore W2951495199C2779304628 @default.
- W2951495199 hasConceptScore W2951495199C2780586882 @default.
- W2951495199 hasConceptScore W2951495199C2781238097 @default.
- W2951495199 hasConceptScore W2951495199C33923547 @default.
- W2951495199 hasConceptScore W2951495199C36289849 @default.
- W2951495199 hasConceptScore W2951495199C41008148 @default.
- W2951495199 hasConceptScore W2951495199C41608201 @default.
- W2951495199 hasConceptScore W2951495199C774472 @default.
- W2951495199 hasConceptScore W2951495199C97931131 @default.
- W2951495199 hasLocation W29514951991 @default.
- W2951495199 hasOpenAccess W2951495199 @default.
- W2951495199 hasPrimaryLocation W29514951991 @default.
- W2951495199 hasRelatedWork W2143564602 @default.
- W2951495199 hasRelatedWork W2806049884 @default.
- W2951495199 hasRelatedWork W2809898047 @default.
- W2951495199 hasRelatedWork W2900969522 @default.