Matches in SemOpenAlex for { <https://semopenalex.org/work/W3208268153> ?p ?o ?g. }
- W3208268153 abstract "Unsupervised representation learning (URL) that learns compact embeddings of high-dimensional data without supervision has achieved remarkable progress recently. Although the ultimate goal of URLs is similar across various scenarios, the related algorithms differ widely in different tasks because they were separately designed according to a specific URL task or data. For example, dimension reduction methods, t-SNE, and UMAP, optimize pair-wise data relationships by preserving the global geometric structure, while self-supervised learning, SimCLR, and BYOL, focus on mining the local statistics of instances under specific augmentations. From a general perspective, we summarize and propose a unified similarity-based URL framework, GenURL, which can adapt to various URL tasks smoothly and efficiently. Based on the manifold assumption, we regard URL tasks as different implicit constraints on the data geometric structure or content that help to seek an optimal low-dimensional representation for the high-dimensional data. Therefore, our method has two key steps to learning task-agnostic representation in URL: (1) data structural modeling and (2) low-dimensional transformation. Specifically, (1) provides a simple yet effective graph-based submodule to model data structures adaptively with predefined or constructed graphs; and based on data-specific pretext tasks, (2) learns compact low-dimensional embeddings. Moreover, (1) and (2) are successfully connected and benefit mutually through our novel objective function. Our comprehensive experiments demonstrate that GenURL achieves consistent state-of-the-art performance in self-supervised visual representation learning, unsupervised knowledge distillation, graph embeddings, and dimension reduction." @default.
- W3208268153 created "2021-11-08" @default.
- W3208268153 creator A5007012841 @default.
- W3208268153 creator A5033050983 @default.
- W3208268153 creator A5055640195 @default.
- W3208268153 creator A5074754590 @default.
- W3208268153 creator A5082786719 @default.
- W3208268153 date "2021-10-27" @default.
- W3208268153 modified "2023-09-27" @default.
- W3208268153 title "GenURL: A General Framework for Unsupervised Representation Learning" @default.
- W3208268153 cites W1821462560 @default.
- W3208268153 cites W2001141328 @default.
- W3208268153 cites W2053186076 @default.
- W3208268153 cites W2112796928 @default.
- W3208268153 cites W2163605009 @default.
- W3208268153 cites W2187089797 @default.
- W3208268153 cites W2194775991 @default.
- W3208268153 cites W2321533354 @default.
- W3208268153 cites W2326925005 @default.
- W3208268153 cites W2519887557 @default.
- W3208268153 cites W2561238782 @default.
- W3208268153 cites W2622263826 @default.
- W3208268153 cites W2739879705 @default.
- W3208268153 cites W2741430497 @default.
- W3208268153 cites W2750384547 @default.
- W3208268153 cites W2786672974 @default.
- W3208268153 cites W2883725317 @default.
- W3208268153 cites W2887783173 @default.
- W3208268153 cites W2943527153 @default.
- W3208268153 cites W2951168573 @default.
- W3208268153 cites W2962742544 @default.
- W3208268153 cites W2963125010 @default.
- W3208268153 cites W2963140444 @default.
- W3208268153 cites W2963163009 @default.
- W3208268153 cites W2964121744 @default.
- W3208268153 cites W2972209102 @default.
- W3208268153 cites W2982242214 @default.
- W3208268153 cites W2990500698 @default.
- W3208268153 cites W2995181141 @default.
- W3208268153 cites W2995607862 @default.
- W3208268153 cites W3005680577 @default.
- W3208268153 cites W3009561768 @default.
- W3208268153 cites W3034381049 @default.
- W3208268153 cites W3034576826 @default.
- W3208268153 cites W3035060554 @default.
- W3208268153 cites W3035524453 @default.
- W3208268153 cites W3042544668 @default.
- W3208268153 cites W3086430417 @default.
- W3208268153 cites W3095121901 @default.
- W3208268153 cites W3104425534 @default.
- W3208268153 cites W3109689953 @default.
- W3208268153 cites W3118608800 @default.
- W3208268153 cites W3123681568 @default.
- W3208268153 cites W3123939835 @default.
- W3208268153 cites W3134652006 @default.
- W3208268153 cites W3171007011 @default.
- W3208268153 cites W343636949 @default.
- W3208268153 cites W2948047609 @default.
- W3208268153 doi "https://doi.org/10.48550/arxiv.2110.14553" @default.
- W3208268153 hasPublicationYear "2021" @default.
- W3208268153 type Work @default.
- W3208268153 sameAs 3208268153 @default.
- W3208268153 citedByCount "0" @default.
- W3208268153 crossrefType "posted-content" @default.
- W3208268153 hasAuthorship W3208268153A5007012841 @default.
- W3208268153 hasAuthorship W3208268153A5033050983 @default.
- W3208268153 hasAuthorship W3208268153A5055640195 @default.
- W3208268153 hasAuthorship W3208268153A5074754590 @default.
- W3208268153 hasAuthorship W3208268153A5082786719 @default.
- W3208268153 hasBestOaLocation W32082681531 @default.
- W3208268153 hasConcept C104317684 @default.
- W3208268153 hasConcept C119857082 @default.
- W3208268153 hasConcept C132525143 @default.
- W3208268153 hasConcept C151876577 @default.
- W3208268153 hasConcept C154945302 @default.
- W3208268153 hasConcept C17744445 @default.
- W3208268153 hasConcept C185592680 @default.
- W3208268153 hasConcept C199539241 @default.
- W3208268153 hasConcept C204241405 @default.
- W3208268153 hasConcept C2776359362 @default.
- W3208268153 hasConcept C41008148 @default.
- W3208268153 hasConcept C55493867 @default.
- W3208268153 hasConcept C59404180 @default.
- W3208268153 hasConcept C70518039 @default.
- W3208268153 hasConcept C8038995 @default.
- W3208268153 hasConcept C80444323 @default.
- W3208268153 hasConcept C94625758 @default.
- W3208268153 hasConceptScore W3208268153C104317684 @default.
- W3208268153 hasConceptScore W3208268153C119857082 @default.
- W3208268153 hasConceptScore W3208268153C132525143 @default.
- W3208268153 hasConceptScore W3208268153C151876577 @default.
- W3208268153 hasConceptScore W3208268153C154945302 @default.
- W3208268153 hasConceptScore W3208268153C17744445 @default.
- W3208268153 hasConceptScore W3208268153C185592680 @default.
- W3208268153 hasConceptScore W3208268153C199539241 @default.
- W3208268153 hasConceptScore W3208268153C204241405 @default.
- W3208268153 hasConceptScore W3208268153C2776359362 @default.
- W3208268153 hasConceptScore W3208268153C41008148 @default.
- W3208268153 hasConceptScore W3208268153C55493867 @default.
- W3208268153 hasConceptScore W3208268153C59404180 @default.