Matches in SemOpenAlex for { <https://semopenalex.org/work/W3157892620> ?p ?o ?g. }
- W3157892620 endingPage "876" @default.
- W3157892620 startingPage "848" @default.
- W3157892620 abstract "In this paper, we propose new learning algorithms for approximating high-dimensional functions using tree tensor networks in a least-squares setting. Given a dimension tree or architecture of the tensor network, we provide an algorithm that generates a sequence of nested tensor subspaces based on a generalization of principal component analysis for multivariate functions. An optimal least-squares method is used for computing projections onto the generated tensor subspaces, using samples generated from a distribution depending on the previously generated subspaces. We provide an error bound in expectation for the obtained approximation. Practical strategies are proposed for adapting the feature spaces and ranks to achieve a prescribed error. Also, we propose an algorithm that progressively constructs the dimension tree by suitable pairings of variables, that allows further reduction of the number of samples necessary to reach that error. Numerical examples illustrate the performance of the proposed algorithms and show that stable approximations are obtained with a number of samples close to the number of free parameters of the estimated tensor networks." @default.
- W3157892620 created "2021-05-10" @default.
- W3157892620 creator A5017081785 @default.
- W3157892620 creator A5031143836 @default.
- W3157892620 creator A5032363972 @default.
- W3157892620 date "2023-07-17" @default.
- W3157892620 modified "2023-09-25" @default.
- W3157892620 title "Active Learning of Tree Tensor Networks using Optimal Least Squares" @default.
- W3157892620 cites W1858056047 @default.
- W3157892620 cites W1967538125 @default.
- W3157892620 cites W1968376936 @default.
- W3157892620 cites W1979098872 @default.
- W3157892620 cites W2001518794 @default.
- W3157892620 cites W2004587738 @default.
- W3157892620 cites W2008657736 @default.
- W3157892620 cites W2013912476 @default.
- W3157892620 cites W2038198231 @default.
- W3157892620 cites W2152828142 @default.
- W3157892620 cites W2160251536 @default.
- W3157892620 cites W2343408237 @default.
- W3157892620 cites W2516041031 @default.
- W3157892620 cites W2597164298 @default.
- W3157892620 cites W2962947392 @default.
- W3157892620 cites W2975849065 @default.
- W3157892620 cites W2995427464 @default.
- W3157892620 cites W3039249709 @default.
- W3157892620 cites W3103888010 @default.
- W3157892620 cites W3206326173 @default.
- W3157892620 cites W4288938614 @default.
- W3157892620 cites W4297907618 @default.
- W3157892620 cites W4323539300 @default.
- W3157892620 doi "https://doi.org/10.1137/21m1415911" @default.
- W3157892620 hasPublicationYear "2023" @default.
- W3157892620 type Work @default.
- W3157892620 sameAs 3157892620 @default.
- W3157892620 citedByCount "1" @default.
- W3157892620 countsByYear W31578926202021 @default.
- W3157892620 crossrefType "journal-article" @default.
- W3157892620 hasAuthorship W3157892620A5017081785 @default.
- W3157892620 hasAuthorship W3157892620A5031143836 @default.
- W3157892620 hasAuthorship W3157892620A5032363972 @default.
- W3157892620 hasBestOaLocation W31578926203 @default.
- W3157892620 hasConcept C105795698 @default.
- W3157892620 hasConcept C113174947 @default.
- W3157892620 hasConcept C11413529 @default.
- W3157892620 hasConcept C114614502 @default.
- W3157892620 hasConcept C12362212 @default.
- W3157892620 hasConcept C126255220 @default.
- W3157892620 hasConcept C134306372 @default.
- W3157892620 hasConcept C154945302 @default.
- W3157892620 hasConcept C155281189 @default.
- W3157892620 hasConcept C177148314 @default.
- W3157892620 hasConcept C185429906 @default.
- W3157892620 hasConcept C202444582 @default.
- W3157892620 hasConcept C27438332 @default.
- W3157892620 hasConcept C2778112365 @default.
- W3157892620 hasConcept C33676613 @default.
- W3157892620 hasConcept C33923547 @default.
- W3157892620 hasConcept C41008148 @default.
- W3157892620 hasConcept C54355233 @default.
- W3157892620 hasConcept C70518039 @default.
- W3157892620 hasConcept C86803240 @default.
- W3157892620 hasConcept C9936470 @default.
- W3157892620 hasConceptScore W3157892620C105795698 @default.
- W3157892620 hasConceptScore W3157892620C113174947 @default.
- W3157892620 hasConceptScore W3157892620C11413529 @default.
- W3157892620 hasConceptScore W3157892620C114614502 @default.
- W3157892620 hasConceptScore W3157892620C12362212 @default.
- W3157892620 hasConceptScore W3157892620C126255220 @default.
- W3157892620 hasConceptScore W3157892620C134306372 @default.
- W3157892620 hasConceptScore W3157892620C154945302 @default.
- W3157892620 hasConceptScore W3157892620C155281189 @default.
- W3157892620 hasConceptScore W3157892620C177148314 @default.
- W3157892620 hasConceptScore W3157892620C185429906 @default.
- W3157892620 hasConceptScore W3157892620C202444582 @default.
- W3157892620 hasConceptScore W3157892620C27438332 @default.
- W3157892620 hasConceptScore W3157892620C2778112365 @default.
- W3157892620 hasConceptScore W3157892620C33676613 @default.
- W3157892620 hasConceptScore W3157892620C33923547 @default.
- W3157892620 hasConceptScore W3157892620C41008148 @default.
- W3157892620 hasConceptScore W3157892620C54355233 @default.
- W3157892620 hasConceptScore W3157892620C70518039 @default.
- W3157892620 hasConceptScore W3157892620C86803240 @default.
- W3157892620 hasConceptScore W3157892620C9936470 @default.
- W3157892620 hasIssue "3" @default.
- W3157892620 hasLocation W31578926201 @default.
- W3157892620 hasLocation W31578926202 @default.
- W3157892620 hasLocation W31578926203 @default.
- W3157892620 hasOpenAccess W3157892620 @default.
- W3157892620 hasPrimaryLocation W31578926201 @default.
- W3157892620 hasRelatedWork W1979158802 @default.
- W3157892620 hasRelatedWork W2058459796 @default.
- W3157892620 hasRelatedWork W2565236036 @default.
- W3157892620 hasRelatedWork W2567092460 @default.
- W3157892620 hasRelatedWork W2951195586 @default.
- W3157892620 hasRelatedWork W3167315549 @default.
- W3157892620 hasRelatedWork W3170528664 @default.
- W3157892620 hasRelatedWork W4287124469 @default.