Matches in SemOpenAlex for { <https://semopenalex.org/work/W4200312714> ?p ?o ?g. }
- W4200312714 endingPage "1148" @default.
- W4200312714 startingPage "1134" @default.
- W4200312714 abstract "The success of deep convolutional networks (ConvNets) generally relies on a massive amount of well-labeled data, which is labor-intensive and time-consuming to collect and annotate in many scenarios. To eliminate such limitation, self-supervised learning (SSL) is recently proposed. Specifically, by solving a pre-designed proxy task, SSL is capable of capturing general-purpose features without requiring human supervision. Existing efforts focus obsessively on designing a particular proxy task but ignore the semanticity of samples that are advantageous to downstream tasks, resulting in the inherent limitation that the learned features are specific to the proxy task, namely the proxy task-specificity of features. In this work, to improve the generalizability of features learned by existing SSL methods, we present a novel self-supervised framework SSL++ to incorporate the proxy task-independent semanticity of samples into the representation learning process. Technically, SSL++ aims to leverage the complementarity, between the low-level generic features learned by a proxy task and the high-level semantic features newly learned by the generated semantic pseudo-labels, to mitigate the task-specificity and improve the generalizability of features. Extensive experiments show that SSL++ performs favorably against the state-of-the-art approaches on the established and latest SSL benchmarks." @default.
- W4200312714 created "2021-12-31" @default.
- W4200312714 creator A5018452576 @default.
- W4200312714 creator A5038083485 @default.
- W4200312714 creator A5047455588 @default.
- W4200312714 creator A5077063251 @default.
- W4200312714 creator A5079361172 @default.
- W4200312714 creator A5089409678 @default.
- W4200312714 date "2022-01-01" @default.
- W4200312714 modified "2023-10-09" @default.
- W4200312714 title "SSL++: Improving Self-Supervised Learning by Mitigating the Proxy Task-Specificity Problem" @default.
- W4200312714 cites W1520997877 @default.
- W4200312714 cites W1903029394 @default.
- W4200312714 cites W2031489346 @default.
- W4200312714 cites W2059515884 @default.
- W4200312714 cites W2117539524 @default.
- W4200312714 cites W2137983211 @default.
- W4200312714 cites W2144796873 @default.
- W4200312714 cites W219040644 @default.
- W4200312714 cites W2198618282 @default.
- W4200312714 cites W2211456655 @default.
- W4200312714 cites W2308529009 @default.
- W4200312714 cites W2321533354 @default.
- W4200312714 cites W2326925005 @default.
- W4200312714 cites W2558661413 @default.
- W4200312714 cites W2575671312 @default.
- W4200312714 cites W2599837529 @default.
- W4200312714 cites W2619697695 @default.
- W4200312714 cites W2769857323 @default.
- W4200312714 cites W2798991696 @default.
- W4200312714 cites W2883725317 @default.
- W4200312714 cites W2913939497 @default.
- W4200312714 cites W2948467580 @default.
- W4200312714 cites W2962824366 @default.
- W4200312714 cites W2962852342 @default.
- W4200312714 cites W2963420272 @default.
- W4200312714 cites W2963465221 @default.
- W4200312714 cites W2963749571 @default.
- W4200312714 cites W2963826423 @default.
- W4200312714 cites W2964060161 @default.
- W4200312714 cites W2979685515 @default.
- W4200312714 cites W2982376094 @default.
- W4200312714 cites W2987741655 @default.
- W4200312714 cites W2990500698 @default.
- W4200312714 cites W2990789488 @default.
- W4200312714 cites W2990873191 @default.
- W4200312714 cites W3014760753 @default.
- W4200312714 cites W3023371261 @default.
- W4200312714 cites W3025966429 @default.
- W4200312714 cites W3034781633 @default.
- W4200312714 cites W3080818206 @default.
- W4200312714 cites W3092095626 @default.
- W4200312714 cites W3098614630 @default.
- W4200312714 cites W3109466111 @default.
- W4200312714 cites W3110446398 @default.
- W4200312714 cites W3113757710 @default.
- W4200312714 cites W3177339245 @default.
- W4200312714 cites W3215810482 @default.
- W4200312714 cites W343636949 @default.
- W4200312714 cites W4292692470 @default.
- W4200312714 doi "https://doi.org/10.1109/tip.2021.3135470" @default.
- W4200312714 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34932477" @default.
- W4200312714 hasPublicationYear "2022" @default.
- W4200312714 type Work @default.
- W4200312714 citedByCount "6" @default.
- W4200312714 countsByYear W42003127142022 @default.
- W4200312714 countsByYear W42003127142023 @default.
- W4200312714 crossrefType "journal-article" @default.
- W4200312714 hasAuthorship W4200312714A5018452576 @default.
- W4200312714 hasAuthorship W4200312714A5038083485 @default.
- W4200312714 hasAuthorship W4200312714A5047455588 @default.
- W4200312714 hasAuthorship W4200312714A5077063251 @default.
- W4200312714 hasAuthorship W4200312714A5079361172 @default.
- W4200312714 hasAuthorship W4200312714A5089409678 @default.
- W4200312714 hasBestOaLocation W42003127142 @default.
- W4200312714 hasConcept C105795698 @default.
- W4200312714 hasConcept C119857082 @default.
- W4200312714 hasConcept C124101348 @default.
- W4200312714 hasConcept C153083717 @default.
- W4200312714 hasConcept C154945302 @default.
- W4200312714 hasConcept C162324750 @default.
- W4200312714 hasConcept C187736073 @default.
- W4200312714 hasConcept C204321447 @default.
- W4200312714 hasConcept C27158222 @default.
- W4200312714 hasConcept C2780148112 @default.
- W4200312714 hasConcept C2780451532 @default.
- W4200312714 hasConcept C33923547 @default.
- W4200312714 hasConcept C41008148 @default.
- W4200312714 hasConcept C81363708 @default.
- W4200312714 hasConcept C94124525 @default.
- W4200312714 hasConceptScore W4200312714C105795698 @default.
- W4200312714 hasConceptScore W4200312714C119857082 @default.
- W4200312714 hasConceptScore W4200312714C124101348 @default.
- W4200312714 hasConceptScore W4200312714C153083717 @default.
- W4200312714 hasConceptScore W4200312714C154945302 @default.
- W4200312714 hasConceptScore W4200312714C162324750 @default.
- W4200312714 hasConceptScore W4200312714C187736073 @default.
- W4200312714 hasConceptScore W4200312714C204321447 @default.