FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2953173407> ?p ?o ?g. }

• W2953173407 abstract "In this paper, we propose a novel multi-task learning method based on the deep convolutional network. The proposed deep network has four convolutional layers, three max-pooling layers, and two parallel fully connected layers. To adjust the deep network to multi-task learning problem, we propose to learn a low-rank deep network so that the relation among different tasks can be explored. We proposed to minimize the number of independent parameter rows of one fully connected layer to explore the relations among different tasks, which is measured by the nuclear norm of the parameter of one fully connected layer, and seek a low-rank parameter matrix. Meanwhile, we also propose to regularize another fully connected layer by sparsity penalty, so that the useful features learned by the lower layers can be selected. The learning problem is solved by an iterative algorithm based on gradient descent and back-propagation algorithms. The proposed algorithm is evaluated over benchmark data sets of multiple face attribute prediction, multi-task natural language processing, and joint economics index predictions. The evaluation results show the advantage of the low-rank deep CNN model over multi-task problems." @default.
• W2953173407 created "2019-06-27" @default.
• W2953173407 creator A5003240878 @default.
• W2953173407 creator A5031857164 @default.
• W2953173407 creator A5072701964 @default.
• W2953173407 date "2019-04-12" @default.
• W2953173407 modified "2023-09-24" @default.
• W2953173407 title "Low-Rank Deep Convolutional Neural Network for Multi-Task Learning" @default.
• W2953173407 cites W1834627138 @default.
• W2953173407 cites W1990395352 @default.
• W2953173407 cites W1997738906 @default.
• W2953173407 cites W2016078102 @default.
• W2953173407 cites W2019654788 @default.
• W2953173407 cites W2026430219 @default.
• W2953173407 cites W2032496386 @default.
• W2953173407 cites W2046405241 @default.
• W2953173407 cites W2076063813 @default.
• W2953173407 cites W2076095133 @default.
• W2953173407 cites W2094035326 @default.
• W2953173407 cites W2109849514 @default.
• W2953173407 cites W2117130368 @default.
• W2953173407 cites W2176950688 @default.
• W2953173407 cites W2184188583 @default.
• W2953173407 cites W22861983 @default.
• W2953173407 cites W2419276856 @default.
• W2953173407 cites W2555827964 @default.
• W2953173407 cites W2597701578 @default.
• W2953173407 cites W2624871570 @default.
• W2953173407 cites W2760843337 @default.
• W2953173407 cites W2765963341 @default.
• W2953173407 cites W2771760728 @default.
• W2953173407 cites W2792189889 @default.
• W2953173407 cites W2794574871 @default.
• W2953173407 cites W2799713005 @default.
• W2953173407 cites W2801440341 @default.
• W2953173407 cites W2802376313 @default.
• W2953173407 cites W2803891953 @default.
• W2953173407 cites W2804709436 @default.
• W2953173407 cites W2807009390 @default.
• W2953173407 cites W2888347392 @default.
• W2953173407 cites W2899218398 @default.
• W2953173407 cites W2904463929 @default.
• W2953173407 cites W2905123482 @default.
• W2953173407 cites W2913340405 @default.
• W2953173407 cites W2916498028 @default.
• W2953173407 cites W2919115771 @default.
• W2953173407 cites W2941129100 @default.
• W2953173407 cites W2952075365 @default.
• W2953173407 doi "https://doi.org/10.48550/arxiv.1904.07320" @default.
• W2953173407 hasPublicationYear "2019" @default.
• W2953173407 type Work @default.
• W2953173407 sameAs 2953173407 @default.
• W2953173407 citedByCount "0" @default.
• W2953173407 crossrefType "posted-content" @default.
• W2953173407 hasAuthorship W2953173407A5003240878 @default.
• W2953173407 hasAuthorship W2953173407A5031857164 @default.
• W2953173407 hasAuthorship W2953173407A5072701964 @default.
• W2953173407 hasBestOaLocation W29531734071 @default.
• W2953173407 hasConcept C108583219 @default.
• W2953173407 hasConcept C114614502 @default.
• W2953173407 hasConcept C119857082 @default.
• W2953173407 hasConcept C13280743 @default.
• W2953173407 hasConcept C153258448 @default.
• W2953173407 hasConcept C154945302 @default.
• W2953173407 hasConcept C162324750 @default.
• W2953173407 hasConcept C164226766 @default.
• W2953173407 hasConcept C185798385 @default.
• W2953173407 hasConcept C187736073 @default.
• W2953173407 hasConcept C205649164 @default.
• W2953173407 hasConcept C2780451532 @default.
• W2953173407 hasConcept C28006648 @default.
• W2953173407 hasConcept C33923547 @default.
• W2953173407 hasConcept C41008148 @default.
• W2953173407 hasConcept C50644808 @default.
• W2953173407 hasConcept C70437156 @default.
• W2953173407 hasConcept C81363708 @default.
• W2953173407 hasConceptScore W2953173407C108583219 @default.
• W2953173407 hasConceptScore W2953173407C114614502 @default.
• W2953173407 hasConceptScore W2953173407C119857082 @default.
• W2953173407 hasConceptScore W2953173407C13280743 @default.
• W2953173407 hasConceptScore W2953173407C153258448 @default.
• W2953173407 hasConceptScore W2953173407C154945302 @default.
• W2953173407 hasConceptScore W2953173407C162324750 @default.
• W2953173407 hasConceptScore W2953173407C164226766 @default.
• W2953173407 hasConceptScore W2953173407C185798385 @default.
• W2953173407 hasConceptScore W2953173407C187736073 @default.
• W2953173407 hasConceptScore W2953173407C205649164 @default.
• W2953173407 hasConceptScore W2953173407C2780451532 @default.
• W2953173407 hasConceptScore W2953173407C28006648 @default.
• W2953173407 hasConceptScore W2953173407C33923547 @default.
• W2953173407 hasConceptScore W2953173407C41008148 @default.
• W2953173407 hasConceptScore W2953173407C50644808 @default.
• W2953173407 hasConceptScore W2953173407C70437156 @default.
• W2953173407 hasConceptScore W2953173407C81363708 @default.
• W2953173407 hasLocation W29531734071 @default.
• W2953173407 hasOpenAccess W2953173407 @default.
• W2953173407 hasPrimaryLocation W29531734071 @default.
• W2953173407 hasRelatedWork W1664573881 @default.
• W2953173407 hasRelatedWork W2337926734 @default.
• W2953173407 hasRelatedWork W2404019942 @default.

• next