FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897786479> ?p ?o ?g. }

• W2897786479 abstract "Autoencoders are unsupervised deep learning models used for learning representations. In literature, autoencoders have shown to perform well on a variety of tasks spread across multiple domains, thereby establishing widespread applicability. Typically, an autoencoder is trained to generate a model that minimizes the reconstruction error between the input and the reconstructed output, computed in terms of the Euclidean distance. While this can be useful for applications related to unsupervised reconstruction, it may not be optimal for classification. In this paper, we propose a novel Supervised COSMOS Autoencoder which utilizes a multi-objective loss function to learn representations that simultaneously encode the (i) between the input and reconstructed vectors in terms of their direction, (ii) distribution of pixel values of the reconstruction with respect to the input sample, while also incorporating (iii) discriminability in the feature learning pipeline. The proposed autoencoder model incorporates a Cosine similarity and Mahalanobis distance based loss function, along with supervision via Mutual Information based loss. Detailed analysis of each component of the proposed model motivates its applicability for feature learning in different classification tasks. The efficacy of Supervised COSMOS autoencoder is demonstrated via extensive experimental evaluations on different image datasets. The proposed model outperforms existing algorithms on MNIST, CIFAR-10, and SVHN databases. It also yields state-of-the-art results on CelebA, LFWA, Adience, and IJB-A databases for attribute prediction and face recognition, respectively." @default.
• W2897786479 created "2018-10-26" @default.
• W2897786479 creator A5011779957 @default.
• W2897786479 creator A5050320880 @default.
• W2897786479 creator A5050521702 @default.
• W2897786479 creator A5066262755 @default.
• W2897786479 creator A5084235091 @default.
• W2897786479 date "2018-10-15" @default.
• W2897786479 modified "2023-09-27" @default.
• W2897786479 title "Supervised COSMOS Autoencoder: Learning Beyond the Euclidean Loss!" @default.
• W2897786479 cites W1499798934 @default.
• W2897786479 cites W1531103298 @default.
• W2897786479 cites W1546200464 @default.
• W2897786479 cites W16016350 @default.
• W2897786479 cites W1834627138 @default.
• W2897786479 cites W1934184906 @default.
• W2897786479 cites W1965804146 @default.
• W2897786479 cites W2002318609 @default.
• W2897786479 cites W2042970394 @default.
• W2897786479 cites W2100495367 @default.
• W2897786479 cites W2104983284 @default.
• W2897786479 cites W2145094598 @default.
• W2897786479 cites W2145607950 @default.
• W2897786479 cites W2178237821 @default.
• W2897786479 cites W2194775991 @default.
• W2897786479 cites W2205500177 @default.
• W2897786479 cites W2218318129 @default.
• W2897786479 cites W2239239723 @default.
• W2897786479 cites W2241794575 @default.
• W2897786479 cites W2248620004 @default.
• W2897786479 cites W2261310161 @default.
• W2897786479 cites W2275811363 @default.
• W2897786479 cites W2294886667 @default.
• W2897786479 cites W2299283369 @default.
• W2897786479 cites W2335728318 @default.
• W2897786479 cites W2337242548 @default.
• W2897786479 cites W2344138609 @default.
• W2897786479 cites W2346735539 @default.
• W2897786479 cites W2401231614 @default.
• W2897786479 cites W2403443505 @default.
• W2897786479 cites W2475024405 @default.
• W2897786479 cites W2528578439 @default.
• W2897786479 cites W2579582149 @default.
• W2897786479 cites W2581613190 @default.
• W2897786479 cites W2605124728 @default.
• W2897786479 cites W2605224169 @default.
• W2897786479 cites W2611289797 @default.
• W2897786479 cites W2731384148 @default.
• W2897786479 cites W2731519763 @default.
• W2897786479 cites W2741295496 @default.
• W2897786479 cites W2806873178 @default.
• W2897786479 cites W2807735086 @default.
• W2897786479 cites W2950277768 @default.
• W2897786479 cites W2951313160 @default.
• W2897786479 cites W2952288113 @default.
• W2897786479 cites W2953039697 @default.
• W2897786479 cites W2962786991 @default.
• W2897786479 cites W2963012631 @default.
• W2897786479 cites W2963244477 @default.
• W2897786479 cites W2963703618 @default.
• W2897786479 cites W2963839617 @default.
• W2897786479 cites W2963919294 @default.
• W2897786479 cites W2964167449 @default.
• W2897786479 cites W3104195492 @default.
• W2897786479 cites W4919037 @default.
• W2897786479 hasPublicationYear "2018" @default.
• W2897786479 type Work @default.
• W2897786479 sameAs 2897786479 @default.
• W2897786479 citedByCount "2" @default.
• W2897786479 countsByYear W28977864792018 @default.
• W2897786479 countsByYear W28977864792020 @default.
• W2897786479 crossrefType "posted-content" @default.
• W2897786479 hasAuthorship W2897786479A5011779957 @default.
• W2897786479 hasAuthorship W2897786479A5050320880 @default.
• W2897786479 hasAuthorship W2897786479A5050521702 @default.
• W2897786479 hasAuthorship W2897786479A5066262755 @default.
• W2897786479 hasAuthorship W2897786479A5084235091 @default.
• W2897786479 hasConcept C101738243 @default.
• W2897786479 hasConcept C103278499 @default.
• W2897786479 hasConcept C108583219 @default.
• W2897786479 hasConcept C115961682 @default.
• W2897786479 hasConcept C119857082 @default.
• W2897786479 hasConcept C120174047 @default.
• W2897786479 hasConcept C136389625 @default.
• W2897786479 hasConcept C138885662 @default.
• W2897786479 hasConcept C153180895 @default.
• W2897786479 hasConcept C154945302 @default.
• W2897786479 hasConcept C162324750 @default.
• W2897786479 hasConcept C176217482 @default.
• W2897786479 hasConcept C190502265 @default.
• W2897786479 hasConcept C1921717 @default.
• W2897786479 hasConcept C21547014 @default.
• W2897786479 hasConcept C2776401178 @default.
• W2897786479 hasConcept C41008148 @default.
• W2897786479 hasConcept C41895202 @default.
• W2897786479 hasConcept C50644808 @default.
• W2897786479 hasConcept C59404180 @default.
• W2897786479 hasConcept C8038995 @default.
• W2897786479 hasConcept C83665646 @default.
• W2897786479 hasConceptScore W2897786479C101738243 @default.

• next