FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3115807029> ?p ?o ?g. }

• W3115807029 endingPage "505" @default.
• W3115807029 startingPage "491" @default.
• W3115807029 abstract "The canonical polyadic decomposition (CPD) allows one to extract compact and interpretable representations of tensors. Several optimization-based methods exist to fit the CPD of a tensor for the standard least-squares (LS) cost function. Extensions have been proposed for more general cost functions such as β-divergences as well. For these non-LS cost functions, a generalized Gauss-Newton (GGN) method has been developed. This is a second-order method that uses an approximation of the Hessian of the cost function to determine the next iterate and with this algorithm, fast convergence can be achieved close to the solution. While it is possible to construct the full Hessian approximation for small tensors, the exact GGN approach becomes too expensive for tensors with larger dimensions, as found in typical applications. In this paper, we therefore propose to use an inexact GGN method and provide several strategies to make this method scalable to large tensors. First, the approximation of the Hessian is only used implicitly and its multilinear structure is exploited during Hessian-vector products, which greatly improves the scalability of the method. Next, we show that by using a compressed instance of the GGN Hessian approximation, the computation time of the inexact GGN method can be lowered even more, with only limited influence on the convergence speed. We also propose dedicated preconditioners for the problem. Further, the maximum likelihood estimator for Rician distributed data is examined in detail as an example of an alternative cost function. This cost function is useful for the analysis of the moduli of complex data, as in functional magnetic resonance imaging, for instance. We compare the proposed method to the existing CPD methods and demonstrate the method's speed and effectiveness on synthetic and simulated real-life data. Finally, we show that the method can scale by using randomized block sampling." @default.
• W3115807029 created "2021-01-05" @default.
• W3115807029 creator A5003192020 @default.
• W3115807029 creator A5009006026 @default.
• W3115807029 creator A5049123030 @default.
• W3115807029 date "2021-04-01" @default.
• W3115807029 modified "2023-10-16" @default.
• W3115807029 title "Inexact Generalized Gauss–Newton for Scaling the Canonical Polyadic Decomposition With Non-Least-Squares Cost Functions" @default.
• W3115807029 cites W1580389772 @default.
• W3115807029 cites W1590929617 @default.
• W3115807029 cites W1963826206 @default.
• W3115807029 cites W1968154520 @default.
• W3115807029 cites W1973246170 @default.
• W3115807029 cites W1974785908 @default.
• W3115807029 cites W1983458887 @default.
• W3115807029 cites W1989811026 @default.
• W3115807029 cites W1990773734 @default.
• W3115807029 cites W1994219736 @default.
• W3115807029 cites W2000389939 @default.
• W3115807029 cites W2005126631 @default.
• W3115807029 cites W2010825468 @default.
• W3115807029 cites W2018850201 @default.
• W3115807029 cites W2022242697 @default.
• W3115807029 cites W2024356620 @default.
• W3115807029 cites W2026034143 @default.
• W3115807029 cites W2038920431 @default.
• W3115807029 cites W2039844283 @default.
• W3115807029 cites W2047680880 @default.
• W3115807029 cites W2054137409 @default.
• W3115807029 cites W2057503509 @default.
• W3115807029 cites W2059784307 @default.
• W3115807029 cites W2066392792 @default.
• W3115807029 cites W2070013413 @default.
• W3115807029 cites W2074523896 @default.
• W3115807029 cites W2088272457 @default.
• W3115807029 cites W2096462613 @default.
• W3115807029 cites W2117756735 @default.
• W3115807029 cites W2119412403 @default.
• W3115807029 cites W2130984546 @default.
• W3115807029 cites W2133665775 @default.
• W3115807029 cites W2133992176 @default.
• W3115807029 cites W2135160607 @default.
• W3115807029 cites W2149192061 @default.
• W3115807029 cites W2151952539 @default.
• W3115807029 cites W2159772265 @default.
• W3115807029 cites W2160047866 @default.
• W3115807029 cites W2164074203 @default.
• W3115807029 cites W2282347063 @default.
• W3115807029 cites W2397636522 @default.
• W3115807029 cites W2469230926 @default.
• W3115807029 cites W2569877296 @default.
• W3115807029 cites W2592422147 @default.
• W3115807029 cites W2924170235 @default.
• W3115807029 cites W2962779982 @default.
• W3115807029 cites W2964024693 @default.
• W3115807029 cites W2970592101 @default.
• W3115807029 cites W3005990752 @default.
• W3115807029 cites W3009014396 @default.
• W3115807029 cites W3012526467 @default.
• W3115807029 cites W3043975730 @default.
• W3115807029 cites W3105254673 @default.
• W3115807029 cites W4320800818 @default.
• W3115807029 cites W49160414 @default.
• W3115807029 doi "https://doi.org/10.1109/jstsp.2020.3045911" @default.
• W3115807029 hasPublicationYear "2021" @default.
• W3115807029 type Work @default.
• W3115807029 sameAs 3115807029 @default.
• W3115807029 citedByCount "3" @default.
• W3115807029 countsByYear W31158070292022 @default.
• W3115807029 countsByYear W31158070292023 @default.
• W3115807029 crossrefType "journal-article" @default.
• W3115807029 hasAuthorship W3115807029A5003192020 @default.
• W3115807029 hasAuthorship W3115807029A5009006026 @default.
• W3115807029 hasAuthorship W3115807029A5049123030 @default.
• W3115807029 hasBestOaLocation W31158070292 @default.
• W3115807029 hasConcept C105795698 @default.
• W3115807029 hasConcept C11413529 @default.
• W3115807029 hasConcept C126255220 @default.
• W3115807029 hasConcept C14036430 @default.
• W3115807029 hasConcept C162324750 @default.
• W3115807029 hasConcept C185429906 @default.
• W3115807029 hasConcept C202444582 @default.
• W3115807029 hasConcept C203616005 @default.
• W3115807029 hasConcept C2777303404 @default.
• W3115807029 hasConcept C28826006 @default.
• W3115807029 hasConcept C33923547 @default.
• W3115807029 hasConcept C45374587 @default.
• W3115807029 hasConcept C50522688 @default.
• W3115807029 hasConcept C78458016 @default.
• W3115807029 hasConcept C84392682 @default.
• W3115807029 hasConcept C86803240 @default.
• W3115807029 hasConceptScore W3115807029C105795698 @default.
• W3115807029 hasConceptScore W3115807029C11413529 @default.
• W3115807029 hasConceptScore W3115807029C126255220 @default.
• W3115807029 hasConceptScore W3115807029C14036430 @default.
• W3115807029 hasConceptScore W3115807029C162324750 @default.
• W3115807029 hasConceptScore W3115807029C185429906 @default.
• W3115807029 hasConceptScore W3115807029C202444582 @default.

• next