FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3203724306> ?p ?o ?g. }

• W3203724306 abstract "Unsupervised disentangled representation learning is a long-standing problem in computer vision. This work proposes a novel framework for performing image clustering from deep embeddings by combining instance-level contrastive learning with a deep embedding based cluster center predictor. Our approach jointly learns representations and predicts cluster centers in an end-to-end manner. This is accomplished via a three-pronged approach that combines a clustering loss, an instance-wise contrastive loss, and an anchor loss. Our fundamental intuition is that using an ensemble loss that incorporates instance-level features and a clustering procedure focusing on semantic similarity reinforces learning better representations in the latent space. We observe that our method performs exceptionally well on popular vision datasets when evaluated using standard clustering metrics such as Normalized Mutual Information (NMI), in addition to producing geometrically well-separated cluster embeddings as defined by the Euclidean distance. Our framework performs on par with widely accepted clustering methods and outperforms the state-of-the-art contrastive learning method on the CIFAR-10 dataset with an NMI score of 0.772, a 7-8% improvement on the strong baseline." @default.
• W3203724306 created "2021-10-11" @default.
• W3203724306 creator A5030601224 @default.
• W3203724306 creator A5049427265 @default.
• W3203724306 creator A5051734156 @default.
• W3203724306 creator A5053891870 @default.
• W3203724306 date "2021-09-26" @default.
• W3203724306 modified "2023-09-27" @default.
• W3203724306 title "Cluster Analysis with Deep Embeddings and Contrastive Learning." @default.
• W3203724306 cites W1507901999 @default.
• W3203724306 cites W1975120776 @default.
• W3203724306 cites W2110798204 @default.
• W3203724306 cites W2127218421 @default.
• W3203724306 cites W2145094598 @default.
• W3203724306 cites W2148349024 @default.
• W3203724306 cites W2152322845 @default.
• W3203724306 cites W2173649752 @default.
• W3203724306 cites W2187089797 @default.
• W3203724306 cites W2321533354 @default.
• W3203724306 cites W2337374958 @default.
• W3203724306 cites W2551604710 @default.
• W3203724306 cites W2608862709 @default.
• W3203724306 cites W2779692282 @default.
• W3203724306 cites W2781711557 @default.
• W3203724306 cites W2788552663 @default.
• W3203724306 cites W2842511635 @default.
• W3203724306 cites W2883594813 @default.
• W3203724306 cites W2883725317 @default.
• W3203724306 cites W2942985259 @default.
• W3203724306 cites W2944828972 @default.
• W3203724306 cites W2949517790 @default.
• W3203724306 cites W2949650786 @default.
• W3203724306 cites W2951292523 @default.
• W3203724306 cites W2951873722 @default.
• W3203724306 cites W2962742544 @default.
• W3203724306 cites W2963684088 @default.
• W3203724306 cites W2964121744 @default.
• W3203724306 cites W2969521304 @default.
• W3203724306 cites W2970971581 @default.
• W3203724306 cites W2971155163 @default.
• W3203724306 cites W2994560339 @default.
• W3203724306 cites W3022061250 @default.
• W3203724306 cites W3034363127 @default.
• W3203724306 cites W3034978746 @default.
• W3203724306 cites W3035060554 @default.
• W3203724306 cites W3035524453 @default.
• W3203724306 cites W3046208551 @default.
• W3203724306 cites W3087124270 @default.
• W3203724306 cites W3095121901 @default.
• W3203724306 cites W3100345210 @default.
• W3203724306 cites W3102419180 @default.
• W3203724306 cites W3103543904 @default.
• W3203724306 cites W3110446398 @default.
• W3203724306 cites W3115293622 @default.
• W3203724306 cites W3122325173 @default.
• W3203724306 cites W3134652006 @default.
• W3203724306 cites W3138310836 @default.
• W3203724306 cites W3151194466 @default.
• W3203724306 cites W3157704640 @default.
• W3203724306 cites W3171007011 @default.
• W3203724306 hasPublicationYear "2021" @default.
• W3203724306 type Work @default.
• W3203724306 sameAs 3203724306 @default.
• W3203724306 citedByCount "1" @default.
• W3203724306 countsByYear W32037243062021 @default.
• W3203724306 crossrefType "posted-content" @default.
• W3203724306 hasAuthorship W3203724306A5030601224 @default.
• W3203724306 hasAuthorship W3203724306A5049427265 @default.
• W3203724306 hasAuthorship W3203724306A5051734156 @default.
• W3203724306 hasAuthorship W3203724306A5053891870 @default.
• W3203724306 hasConcept C103278499 @default.
• W3203724306 hasConcept C104047586 @default.
• W3203724306 hasConcept C108583219 @default.
• W3203724306 hasConcept C111472728 @default.
• W3203724306 hasConcept C115961682 @default.
• W3203724306 hasConcept C119857082 @default.
• W3203724306 hasConcept C132010649 @default.
• W3203724306 hasConcept C138885662 @default.
• W3203724306 hasConcept C153180895 @default.
• W3203724306 hasConcept C154945302 @default.
• W3203724306 hasConcept C17744445 @default.
• W3203724306 hasConcept C186767784 @default.
• W3203724306 hasConcept C199539241 @default.
• W3203724306 hasConcept C2776359362 @default.
• W3203724306 hasConcept C41008148 @default.
• W3203724306 hasConcept C41608201 @default.
• W3203724306 hasConcept C59404180 @default.
• W3203724306 hasConcept C73555534 @default.
• W3203724306 hasConcept C94625758 @default.
• W3203724306 hasConcept C94641424 @default.
• W3203724306 hasConceptScore W3203724306C103278499 @default.
• W3203724306 hasConceptScore W3203724306C104047586 @default.
• W3203724306 hasConceptScore W3203724306C108583219 @default.
• W3203724306 hasConceptScore W3203724306C111472728 @default.
• W3203724306 hasConceptScore W3203724306C115961682 @default.
• W3203724306 hasConceptScore W3203724306C119857082 @default.
• W3203724306 hasConceptScore W3203724306C132010649 @default.
• W3203724306 hasConceptScore W3203724306C138885662 @default.
• W3203724306 hasConceptScore W3203724306C153180895 @default.
• W3203724306 hasConceptScore W3203724306C154945302 @default.

• next