FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2974544900> ?p ?o ?g. }

• W2974544900 endingPage "872" @default.
• W2974544900 startingPage "858" @default.
• W2974544900 abstract "Multimodal learning aims to discover the relationship between multiple modalities. It has become an important research topic due to extensive multimodal applications such as cross-modal retrieval. This paper attempts to address the modality heterogeneity problem based on Gaussian process latent variable models (GPLVMs) to represent multimodal data in a common space. Previous multimodal GPLVM extensions generally adopt individual learning schemes on latent representations and kernel hyperparameters, which ignore their intrinsic relationship. To exploit strong complementarity among different modalities and GPLVM components, we develop a novel learning scheme called Harmonization, where latent model parameters are jointly learned from each other. Beyond the correlation fitting or intra-modal structure preservation paradigms widely used in existing studies, the harmonization is derived in a model-driven manner to encourage the agreement between modality-specific GP kernels and the similarity of latent representations. We present a range of multimodal learning models by incorporating the harmonization mechanism into several representative GPLVM-based approaches. Experimental results on four benchmark datasets show that the proposed models outperform the strong baselines for cross-modal retrieval tasks, and that the harmonized multimodal learning method is superior in discovering semantically consistent latent representation." @default.
• W2974544900 created "2019-09-26" @default.
• W2974544900 creator A5008752777 @default.
• W2974544900 creator A5028597017 @default.
• W2974544900 creator A5034112897 @default.
• W2974544900 creator A5047455588 @default.
• W2974544900 date "2021-03-01" @default.
• W2974544900 modified "2023-10-17" @default.
• W2974544900 title "Harmonized Multimodal Learning with Gaussian Process Latent Variable Models" @default.
• W2974544900 cites W1493404174 @default.
• W2974544900 cites W1592510033 @default.
• W2974544900 cites W1895577753 @default.
• W2974544900 cites W1916445035 @default.
• W2974544900 cites W1949478088 @default.
• W2974544900 cites W2022398331 @default.
• W2974544900 cites W2029163572 @default.
• W2974544900 cites W2051812123 @default.
• W2974544900 cites W2061223839 @default.
• W2974544900 cites W2062955551 @default.
• W2974544900 cites W2064797228 @default.
• W2974544900 cites W2070753207 @default.
• W2974544900 cites W2071207147 @default.
• W2974544900 cites W2076462394 @default.
• W2974544900 cites W2100235303 @default.
• W2974544900 cites W2104563967 @default.
• W2974544900 cites W2106277773 @default.
• W2974544900 cites W2120405375 @default.
• W2974544900 cites W2133294540 @default.
• W2974544900 cites W2138118304 @default.
• W2974544900 cites W2171461439 @default.
• W2974544900 cites W2210322478 @default.
• W2974544900 cites W2211092169 @default.
• W2974544900 cites W2218644571 @default.
• W2974544900 cites W2266728343 @default.
• W2974544900 cites W2410942542 @default.
• W2974544900 cites W2564898401 @default.
• W2974544900 cites W2605649771 @default.
• W2974544900 cites W2613677041 @default.
• W2974544900 cites W2619383789 @default.
• W2974544900 cites W2623201052 @default.
• W2974544900 cites W2748725735 @default.
• W2974544900 cites W2765178306 @default.
• W2974544900 cites W2776110275 @default.
• W2974544900 cites W2896798564 @default.
• W2974544900 cites W3098232083 @default.
• W2974544900 cites W4213009331 @default.
• W2974544900 cites W4237723258 @default.
• W2974544900 cites W4248892431 @default.
• W2974544900 doi "https://doi.org/10.1109/tpami.2019.2942028" @default.
• W2974544900 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31545710" @default.
• W2974544900 hasPublicationYear "2021" @default.
• W2974544900 type Work @default.
• W2974544900 sameAs 2974544900 @default.
• W2974544900 citedByCount "12" @default.
• W2974544900 countsByYear W29745449002020 @default.
• W2974544900 countsByYear W29745449002021 @default.
• W2974544900 countsByYear W29745449002022 @default.
• W2974544900 countsByYear W29745449002023 @default.
• W2974544900 crossrefType "journal-article" @default.
• W2974544900 hasAuthorship W2974544900A5008752777 @default.
• W2974544900 hasAuthorship W2974544900A5028597017 @default.
• W2974544900 hasAuthorship W2974544900A5034112897 @default.
• W2974544900 hasAuthorship W2974544900A5047455588 @default.
• W2974544900 hasBestOaLocation W29745449002 @default.
• W2974544900 hasConcept C111919701 @default.
• W2974544900 hasConcept C119857082 @default.
• W2974544900 hasConcept C121332964 @default.
• W2974544900 hasConcept C134306372 @default.
• W2974544900 hasConcept C153180895 @default.
• W2974544900 hasConcept C154945302 @default.
• W2974544900 hasConcept C163716315 @default.
• W2974544900 hasConcept C182365436 @default.
• W2974544900 hasConcept C33923547 @default.
• W2974544900 hasConcept C41008148 @default.
• W2974544900 hasConcept C51167844 @default.
• W2974544900 hasConcept C61326573 @default.
• W2974544900 hasConcept C62520636 @default.
• W2974544900 hasConcept C98045186 @default.
• W2974544900 hasConceptScore W2974544900C111919701 @default.
• W2974544900 hasConceptScore W2974544900C119857082 @default.
• W2974544900 hasConceptScore W2974544900C121332964 @default.
• W2974544900 hasConceptScore W2974544900C134306372 @default.
• W2974544900 hasConceptScore W2974544900C153180895 @default.
• W2974544900 hasConceptScore W2974544900C154945302 @default.
• W2974544900 hasConceptScore W2974544900C163716315 @default.
• W2974544900 hasConceptScore W2974544900C182365436 @default.
• W2974544900 hasConceptScore W2974544900C33923547 @default.
• W2974544900 hasConceptScore W2974544900C41008148 @default.
• W2974544900 hasConceptScore W2974544900C51167844 @default.
• W2974544900 hasConceptScore W2974544900C61326573 @default.
• W2974544900 hasConceptScore W2974544900C62520636 @default.
• W2974544900 hasConceptScore W2974544900C98045186 @default.
• W2974544900 hasFunder F4320321001 @default.
• W2974544900 hasFunder F4320321543 @default.
• W2974544900 hasFunder F4320335777 @default.
• W2974544900 hasIssue "3" @default.
• W2974544900 hasLocation W29745449001 @default.
• W2974544900 hasLocation W29745449002 @default.

• next