FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3096189675> ?p ?o ?g. }

• W3096189675 abstract "Self-supervised learning aims to learn good representations with unlabeled data. Recent works have shown that larger models benefit more from self-supervised learning than smaller models. As a result, the gap between supervised and self-supervised learning has been greatly reduced for larger models. In this work, instead of designing a new pseudo task for self-supervised learning, we develop a model compression method to compress an already learned, deep self-supervised model (teacher) to a smaller one (student). We train the student model so that it mimics the relative similarity between the data points in the teacher's embedding space. For AlexNet, our method outperforms all previous methods including the fully supervised model on ImageNet linear evaluation (59.0% compared to 56.5%) and on nearest neighbor evaluation (50.7% compared to 41.4%). To the best of our knowledge, this is the first time a self-supervised AlexNet has outperformed supervised one on ImageNet classification. Our code is available here: this https URL" @default.
• W3096189675 created "2020-11-09" @default.
• W3096189675 creator A5031794370 @default.
• W3096189675 creator A5031860916 @default.
• W3096189675 creator A5084289151 @default.
• W3096189675 date "2020-10-28" @default.
• W3096189675 modified "2023-09-27" @default.
• W3096189675 title "CompRess: Self-Supervised Learning by Compressing Representations" @default.
• W3096189675 cites W1536680647 @default.
• W3096189675 cites W1821462560 @default.
• W3096189675 cites W1989684337 @default.
• W3096189675 cites W2031489346 @default.
• W3096189675 cites W2117539524 @default.
• W3096189675 cites W2134670479 @default.
• W3096189675 cites W2134797427 @default.
• W3096189675 cites W2138011018 @default.
• W3096189675 cites W2138621090 @default.
• W3096189675 cites W2148349024 @default.
• W3096189675 cites W2222512263 @default.
• W3096189675 cites W2294370754 @default.
• W3096189675 cites W2321533354 @default.
• W3096189675 cites W2326925005 @default.
• W3096189675 cites W2558661413 @default.
• W3096189675 cites W2561238782 @default.
• W3096189675 cites W2592691248 @default.
• W3096189675 cites W2739879705 @default.
• W3096189675 cites W2750549109 @default.
• W3096189675 cites W2798680770 @default.
• W3096189675 cites W2798991696 @default.
• W3096189675 cites W2802198257 @default.
• W3096189675 cites W2842511635 @default.
• W3096189675 cites W2883725317 @default.
• W3096189675 cites W2887783173 @default.
• W3096189675 cites W2887997457 @default.
• W3096189675 cites W2935908327 @default.
• W3096189675 cites W2936864631 @default.
• W3096189675 cites W2941964676 @default.
• W3096189675 cites W2944828972 @default.
• W3096189675 cites W2948467580 @default.
• W3096189675 cites W2948582784 @default.
• W3096189675 cites W2949517790 @default.
• W3096189675 cites W2962742544 @default.
• W3096189675 cites W2963140444 @default.
• W3096189675 cites W2963265008 @default.
• W3096189675 cites W2963420272 @default.
• W3096189675 cites W2963465221 @default.
• W3096189675 cites W2963534679 @default.
• W3096189675 cites W2964005298 @default.
• W3096189675 cites W2964118293 @default.
• W3096189675 cites W2964222566 @default.
• W3096189675 cites W2971155163 @default.
• W3096189675 cites W2982242214 @default.
• W3096189675 cites W2982376094 @default.
• W3096189675 cites W2986015886 @default.
• W3096189675 cites W2990500698 @default.
• W3096189675 cites W2990873191 @default.
• W3096189675 cites W2994536315 @default.
• W3096189675 cites W2995181141 @default.
• W3096189675 cites W2995607862 @default.
• W3096189675 cites W3005680577 @default.
• W3096189675 cites W3009561768 @default.
• W3096189675 cites W3026092005 @default.
• W3096189675 cites W3034345981 @default.
• W3096189675 cites W3035060554 @default.
• W3096189675 cites W3035524453 @default.
• W3096189675 cites W3036224891 @default.
• W3096189675 cites W343636949 @default.
• W3096189675 cites W14333344 @default.
• W3096189675 hasPublicationYear "2020" @default.
• W3096189675 type Work @default.
• W3096189675 sameAs 3096189675 @default.
• W3096189675 citedByCount "2" @default.
• W3096189675 countsByYear W30961896752021 @default.
• W3096189675 crossrefType "posted-content" @default.
• W3096189675 hasAuthorship W3096189675A5031794370 @default.
• W3096189675 hasAuthorship W3096189675A5031860916 @default.
• W3096189675 hasAuthorship W3096189675A5084289151 @default.
• W3096189675 hasConcept C103278499 @default.
• W3096189675 hasConcept C111919701 @default.
• W3096189675 hasConcept C113238511 @default.
• W3096189675 hasConcept C115961682 @default.
• W3096189675 hasConcept C119857082 @default.
• W3096189675 hasConcept C136389625 @default.
• W3096189675 hasConcept C153180895 @default.
• W3096189675 hasConcept C154945302 @default.
• W3096189675 hasConcept C2778572836 @default.
• W3096189675 hasConcept C41008148 @default.
• W3096189675 hasConcept C41608201 @default.
• W3096189675 hasConcept C50644808 @default.
• W3096189675 hasConcept C58973888 @default.
• W3096189675 hasConceptScore W3096189675C103278499 @default.
• W3096189675 hasConceptScore W3096189675C111919701 @default.
• W3096189675 hasConceptScore W3096189675C113238511 @default.
• W3096189675 hasConceptScore W3096189675C115961682 @default.
• W3096189675 hasConceptScore W3096189675C119857082 @default.
• W3096189675 hasConceptScore W3096189675C136389625 @default.
• W3096189675 hasConceptScore W3096189675C153180895 @default.
• W3096189675 hasConceptScore W3096189675C154945302 @default.
• W3096189675 hasConceptScore W3096189675C2778572836 @default.
• W3096189675 hasConceptScore W3096189675C41008148 @default.

• next