Matches in SemOpenAlex for { <https://semopenalex.org/work/W4381952024> ?p ?o ?g. }
- W4381952024 endingPage "100479" @default.
- W4381952024 startingPage "100479" @default.
- W4381952024 abstract "In this paper, we study the problem of multilinear multitask learning (MLMTL), in which all tasks are stacked into a third-order tensor for consideration. In contrast to conventional multitask learning, MLMTL can explore inherent correlations among multiple tasks in a better manner by utilizing multilinear low rank structure. Existing approaches about MLMTL are mainly based on the sum of singular values for approximating low rank matrices obtained by matricizing the third-order tensor. However, these methods are suboptimal in the Tucker rank approximation. In order to elucidate intrinsic correlations among multiple tasks, we present a new approach by the use of transformed tensor nuclear norm (TTNN) constraint in the objective function. The main advantage of the proposed approach is that it can acquire a low transformed multi-rank structure in a transformed tensor by applying suitable unitary transformations which is helpful to determine principal components in grouping multiple tasks for describing their intrinsic correlations more precisely. Furthermore, we establish an excess risk bound of the minimizer of the proposed TTNN approach. Experimental results including synthetic problems and real-world images, show that the mean-square errors of the proposed method is lower than those of the existing methods for different number of tasks and training samples in MLMTL." @default.
- W4381952024 created "2023-06-26" @default.
- W4381952024 creator A5018117569 @default.
- W4381952024 creator A5029441139 @default.
- W4381952024 creator A5072740546 @default.
- W4381952024 date "2023-09-01" @default.
- W4381952024 modified "2023-10-01" @default.
- W4381952024 title "Multilinear multitask learning by transformed tensor singular value decomposition" @default.
- W4381952024 cites W1963826206 @default.
- W4381952024 cites W1967849786 @default.
- W4381952024 cites W1967901188 @default.
- W4381952024 cites W2018096278 @default.
- W4381952024 cites W2024165284 @default.
- W4381952024 cites W2043571470 @default.
- W4381952024 cites W2045704273 @default.
- W4381952024 cites W2056145308 @default.
- W4381952024 cites W2065180801 @default.
- W4381952024 cites W2083042020 @default.
- W4381952024 cites W2102630986 @default.
- W4381952024 cites W2108495711 @default.
- W4381952024 cites W2110332320 @default.
- W4381952024 cites W2139208832 @default.
- W4381952024 cites W2141846515 @default.
- W4381952024 cites W2143104527 @default.
- W4381952024 cites W2143419558 @default.
- W4381952024 cites W2331050942 @default.
- W4381952024 cites W2394550695 @default.
- W4381952024 cites W2669889762 @default.
- W4381952024 cites W2753709519 @default.
- W4381952024 cites W2913340405 @default.
- W4381952024 cites W2953742623 @default.
- W4381952024 cites W2963885538 @default.
- W4381952024 cites W2997421030 @default.
- W4381952024 cites W3011789951 @default.
- W4381952024 cites W3014517104 @default.
- W4381952024 cites W3100110387 @default.
- W4381952024 cites W3104240813 @default.
- W4381952024 cites W3120378560 @default.
- W4381952024 cites W3141797743 @default.
- W4381952024 cites W3168230249 @default.
- W4381952024 cites W4233413206 @default.
- W4381952024 cites W4300179783 @default.
- W4381952024 cites W4362474484 @default.
- W4381952024 doi "https://doi.org/10.1016/j.mlwa.2023.100479" @default.
- W4381952024 hasPublicationYear "2023" @default.
- W4381952024 type Work @default.
- W4381952024 citedByCount "0" @default.
- W4381952024 crossrefType "journal-article" @default.
- W4381952024 hasAuthorship W4381952024A5018117569 @default.
- W4381952024 hasAuthorship W4381952024A5029441139 @default.
- W4381952024 hasAuthorship W4381952024A5072740546 @default.
- W4381952024 hasBestOaLocation W43819520241 @default.
- W4381952024 hasConcept C109282560 @default.
- W4381952024 hasConcept C11413529 @default.
- W4381952024 hasConcept C114614502 @default.
- W4381952024 hasConcept C121332964 @default.
- W4381952024 hasConcept C126255220 @default.
- W4381952024 hasConcept C14036430 @default.
- W4381952024 hasConcept C154945302 @default.
- W4381952024 hasConcept C155281189 @default.
- W4381952024 hasConcept C158693339 @default.
- W4381952024 hasConcept C164226766 @default.
- W4381952024 hasConcept C17744445 @default.
- W4381952024 hasConcept C199539241 @default.
- W4381952024 hasConcept C202444582 @default.
- W4381952024 hasConcept C22789450 @default.
- W4381952024 hasConcept C2524010 @default.
- W4381952024 hasConcept C2776036281 @default.
- W4381952024 hasConcept C33923547 @default.
- W4381952024 hasConcept C41008148 @default.
- W4381952024 hasConcept C62520636 @default.
- W4381952024 hasConcept C67820243 @default.
- W4381952024 hasConcept C78458016 @default.
- W4381952024 hasConcept C84392682 @default.
- W4381952024 hasConcept C86803240 @default.
- W4381952024 hasConcept C92207270 @default.
- W4381952024 hasConceptScore W4381952024C109282560 @default.
- W4381952024 hasConceptScore W4381952024C11413529 @default.
- W4381952024 hasConceptScore W4381952024C114614502 @default.
- W4381952024 hasConceptScore W4381952024C121332964 @default.
- W4381952024 hasConceptScore W4381952024C126255220 @default.
- W4381952024 hasConceptScore W4381952024C14036430 @default.
- W4381952024 hasConceptScore W4381952024C154945302 @default.
- W4381952024 hasConceptScore W4381952024C155281189 @default.
- W4381952024 hasConceptScore W4381952024C158693339 @default.
- W4381952024 hasConceptScore W4381952024C164226766 @default.
- W4381952024 hasConceptScore W4381952024C17744445 @default.
- W4381952024 hasConceptScore W4381952024C199539241 @default.
- W4381952024 hasConceptScore W4381952024C202444582 @default.
- W4381952024 hasConceptScore W4381952024C22789450 @default.
- W4381952024 hasConceptScore W4381952024C2524010 @default.
- W4381952024 hasConceptScore W4381952024C2776036281 @default.
- W4381952024 hasConceptScore W4381952024C33923547 @default.
- W4381952024 hasConceptScore W4381952024C41008148 @default.
- W4381952024 hasConceptScore W4381952024C62520636 @default.
- W4381952024 hasConceptScore W4381952024C67820243 @default.
- W4381952024 hasConceptScore W4381952024C78458016 @default.
- W4381952024 hasConceptScore W4381952024C84392682 @default.