FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3109306989> ?p ?o ?g. }

• W3109306989 abstract "Most existing distance metric learning approaches use fully labeled data to learn the sample similarities in an embedding space. We present a self-training framework, SLADE, to improve retrieval performance by leveraging additional unlabeled data. We first train a teacher model on the labeled data and use it to generate pseudo labels for the unlabeled data. We then train a student model on both labels and pseudo labels to generate final feature embeddings. We use self-supervised representation learning to initialize the teacher model. To better deal with noisy pseudo labels generated by the teacher network, we design a new feature basis learning component for the student network, which learns basis functions of feature representations for unlabeled data. The learned basis vectors better measure the pairwise similarity and are used to select high-confident samples for training the student network. We evaluate our method on standard retrieval benchmarks: CUB-200, Cars-196 and In-shop. Experimental results demonstrate that our approach significantly improves the performance over the state-of-the-art methods." @default.
• W3109306989 created "2020-12-07" @default.
• W3109306989 creator A5008915336 @default.
• W3109306989 creator A5051293560 @default.
• W3109306989 creator A5055264952 @default.
• W3109306989 creator A5065995768 @default.
• W3109306989 creator A5078602610 @default.
• W3109306989 date "2020-11-20" @default.
• W3109306989 modified "2023-09-27" @default.
• W3109306989 title "SLADE: A Self-Training Framework For Distance Metric Learning" @default.
• W3109306989 cites W1797268635 @default.
• W3109306989 cites W1821462560 @default.
• W3109306989 cites W1954152232 @default.
• W3109306989 cites W1958236864 @default.
• W3109306989 cites W2106053110 @default.
• W3109306989 cites W2108598243 @default.
• W3109306989 cites W2138011018 @default.
• W3109306989 cites W2138621090 @default.
• W3109306989 cites W2471768434 @default.
• W3109306989 cites W2605102252 @default.
• W3109306989 cites W2883725317 @default.
• W3109306989 cites W2924876209 @default.
• W3109306989 cites W2953214814 @default.
• W3109306989 cites W2953271441 @default.
• W3109306989 cites W2962898354 @default.
• W3109306989 cites W2963026686 @default.
• W3109306989 cites W2964211610 @default.
• W3109306989 cites W2964271799 @default.
• W3109306989 cites W3009561768 @default.
• W3109306989 cites W3012668059 @default.
• W3109306989 cites W3022509031 @default.
• W3109306989 cites W3034202663 @default.
• W3109306989 cites W3034576826 @default.
• W3109306989 cites W3034756453 @default.
• W3109306989 cites W3034978746 @default.
• W3109306989 cites W3035003500 @default.
• W3109306989 cites W3035014997 @default.
• W3109306989 cites W3035060554 @default.
• W3109306989 cites W3035160371 @default.
• W3109306989 cites W3035524453 @default.
• W3109306989 cites W3081378822 @default.
• W3109306989 cites W3095121901 @default.
• W3109306989 cites W3098656025 @default.
• W3109306989 cites W3100859887 @default.
• W3109306989 cites W3103152812 @default.
• W3109306989 cites W3106778652 @default.
• W3109306989 cites W3109684201 @default.
• W3109306989 cites W3109689953 @default.
• W3109306989 doi "https://doi.org/10.48550/arxiv.2011.10269" @default.
• W3109306989 hasPublicationYear "2020" @default.
• W3109306989 type Work @default.
• W3109306989 sameAs 3109306989 @default.
• W3109306989 citedByCount "0" @default.
• W3109306989 crossrefType "posted-content" @default.
• W3109306989 hasAuthorship W3109306989A5008915336 @default.
• W3109306989 hasAuthorship W3109306989A5051293560 @default.
• W3109306989 hasAuthorship W3109306989A5055264952 @default.
• W3109306989 hasAuthorship W3109306989A5065995768 @default.
• W3109306989 hasAuthorship W3109306989A5078602610 @default.
• W3109306989 hasBestOaLocation W31093069891 @default.
• W3109306989 hasConcept C103278499 @default.
• W3109306989 hasConcept C115961682 @default.
• W3109306989 hasConcept C119857082 @default.
• W3109306989 hasConcept C121332964 @default.
• W3109306989 hasConcept C12426560 @default.
• W3109306989 hasConcept C138885662 @default.
• W3109306989 hasConcept C153180895 @default.
• W3109306989 hasConcept C154945302 @default.
• W3109306989 hasConcept C162324750 @default.
• W3109306989 hasConcept C168167062 @default.
• W3109306989 hasConcept C176217482 @default.
• W3109306989 hasConcept C17744445 @default.
• W3109306989 hasConcept C184898388 @default.
• W3109306989 hasConcept C185592680 @default.
• W3109306989 hasConcept C198531522 @default.
• W3109306989 hasConcept C199539241 @default.
• W3109306989 hasConcept C21547014 @default.
• W3109306989 hasConcept C2524010 @default.
• W3109306989 hasConcept C2776359362 @default.
• W3109306989 hasConcept C2776401178 @default.
• W3109306989 hasConcept C33923547 @default.
• W3109306989 hasConcept C41008148 @default.
• W3109306989 hasConcept C41608201 @default.
• W3109306989 hasConcept C41895202 @default.
• W3109306989 hasConcept C43617362 @default.
• W3109306989 hasConcept C59404180 @default.
• W3109306989 hasConcept C83665646 @default.
• W3109306989 hasConcept C94625758 @default.
• W3109306989 hasConcept C97355855 @default.
• W3109306989 hasConceptScore W3109306989C103278499 @default.
• W3109306989 hasConceptScore W3109306989C115961682 @default.
• W3109306989 hasConceptScore W3109306989C119857082 @default.
• W3109306989 hasConceptScore W3109306989C121332964 @default.
• W3109306989 hasConceptScore W3109306989C12426560 @default.
• W3109306989 hasConceptScore W3109306989C138885662 @default.
• W3109306989 hasConceptScore W3109306989C153180895 @default.
• W3109306989 hasConceptScore W3109306989C154945302 @default.
• W3109306989 hasConceptScore W3109306989C162324750 @default.
• W3109306989 hasConceptScore W3109306989C168167062 @default.
• W3109306989 hasConceptScore W3109306989C176217482 @default.

• next