FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2023062008> ?p ?o ?g. }

• W2023062008 endingPage "1916" @default.
• W2023062008 startingPage "1904" @default.
• W2023062008 abstract "Nonnegative matrix factorization (NMF) solves the following problem: find nonnegative matrices A∈R+M×R and X∈R+R×T such that Y≅AX, given only Y∈RM×T and the assigned index R. This method has found a wide spectrum of applications in signal and image processing, such as blind source separation (BSS), spectra recovering, pattern recognition, segmentation or clustering. Such a factorization is usually performed with an alternating gradient descent technique that is applied to the squared Euclidean distance or Kullback–Leibler divergence. This approach has been used in the widely known Lee–Seung NMF algorithms that belong to a class of multiplicative iterative algorithms. It is well known that these algorithms, in spite of their low complexity, are slowly convergent, give only a strictly positive solution, and can easily fall into local minima of a nonconvex cost function. In this paper, we propose to take advantage of the second-order terms of a cost function to overcome the disadvantages of gradient (multiplicative) algorithms. First, a projected quasi-Newton method is presented, where a regularized Hessian with the Levenberg–Marquardt approach is inverted with the Q-less QR decomposition. Since the matrices A and/or X are usually sparse, a more sophisticated hybrid approach based on the gradient projection conjugate gradient (GPCG) algorithm, which was invented by More and Toraldo, is adapted for NMF. The gradient projection (GP) method is exploited to find zero-value components (active), and then the Newton steps are taken only to compute positive components (inactive) with the conjugate gradient (CG) method. As a cost function, we used the α-divergence that unifies many well-known cost functions. We applied our new NMF method to a BSS problem with mixed signals and images. The results demonstrate the high robustness of our method." @default.
• W2023062008 created "2016-06-24" @default.
• W2023062008 creator A5018676117 @default.
• W2023062008 creator A5022656662 @default.
• W2023062008 date "2007-08-01" @default.
• W2023062008 modified "2023-09-26" @default.
• W2023062008 title "Nonnegative matrix factorization with constrained second-order optimization" @default.
• W2023062008 cites W1790954942 @default.
• W2023062008 cites W1902027874 @default.
• W2023062008 cites W1969182276 @default.
• W2023062008 cites W1991380130 @default.
• W2023062008 cites W2017288758 @default.
• W2023062008 cites W2026289804 @default.
• W2023062008 cites W2053566983 @default.
• W2023062008 cites W2082445317 @default.
• W2023062008 cites W2088909704 @default.
• W2023062008 cites W2100493539 @default.
• W2023062008 cites W2113359929 @default.
• W2023062008 cites W2116147316 @default.
• W2023062008 cites W2134306264 @default.
• W2023062008 cites W2135384781 @default.
• W2023062008 cites W2136637647 @default.
• W2023062008 cites W2136787567 @default.
• W2023062008 cites W2137573778 @default.
• W2023062008 cites W2146913572 @default.
• W2023062008 cites W2151517435 @default.
• W2023062008 cites W2152320972 @default.
• W2023062008 cites W2162409952 @default.
• W2023062008 cites W2170608748 @default.
• W2023062008 cites W2316564661 @default.
• W2023062008 cites W2749035964 @default.
• W2023062008 doi "https://doi.org/10.1016/j.sigpro.2007.01.024" @default.
• W2023062008 hasPublicationYear "2007" @default.
• W2023062008 type Work @default.
• W2023062008 sameAs 2023062008 @default.
• W2023062008 citedByCount "89" @default.
• W2023062008 countsByYear W20230620082012 @default.
• W2023062008 countsByYear W20230620082013 @default.
• W2023062008 countsByYear W20230620082014 @default.
• W2023062008 countsByYear W20230620082015 @default.
• W2023062008 countsByYear W20230620082016 @default.
• W2023062008 countsByYear W20230620082017 @default.
• W2023062008 countsByYear W20230620082018 @default.
• W2023062008 countsByYear W20230620082019 @default.
• W2023062008 countsByYear W20230620082020 @default.
• W2023062008 countsByYear W20230620082021 @default.
• W2023062008 countsByYear W20230620082022 @default.
• W2023062008 countsByYear W20230620082023 @default.
• W2023062008 crossrefType "journal-article" @default.
• W2023062008 hasAuthorship W2023062008A5018676117 @default.
• W2023062008 hasAuthorship W2023062008A5022656662 @default.
• W2023062008 hasConcept C11413529 @default.
• W2023062008 hasConcept C114614502 @default.
• W2023062008 hasConcept C115680565 @default.
• W2023062008 hasConcept C120317606 @default.
• W2023062008 hasConcept C121332964 @default.
• W2023062008 hasConcept C127162648 @default.
• W2023062008 hasConcept C134306372 @default.
• W2023062008 hasConcept C152671427 @default.
• W2023062008 hasConcept C153258448 @default.
• W2023062008 hasConcept C154945302 @default.
• W2023062008 hasConcept C158693339 @default.
• W2023062008 hasConcept C186633575 @default.
• W2023062008 hasConcept C203616005 @default.
• W2023062008 hasConcept C28826006 @default.
• W2023062008 hasConcept C31258907 @default.
• W2023062008 hasConcept C33923547 @default.
• W2023062008 hasConcept C41008148 @default.
• W2023062008 hasConcept C42355184 @default.
• W2023062008 hasConcept C42747912 @default.
• W2023062008 hasConcept C50644808 @default.
• W2023062008 hasConcept C57493831 @default.
• W2023062008 hasConcept C62520636 @default.
• W2023062008 hasConcept C81184566 @default.
• W2023062008 hasConceptScore W2023062008C11413529 @default.
• W2023062008 hasConceptScore W2023062008C114614502 @default.
• W2023062008 hasConceptScore W2023062008C115680565 @default.
• W2023062008 hasConceptScore W2023062008C120317606 @default.
• W2023062008 hasConceptScore W2023062008C121332964 @default.
• W2023062008 hasConceptScore W2023062008C127162648 @default.
• W2023062008 hasConceptScore W2023062008C134306372 @default.
• W2023062008 hasConceptScore W2023062008C152671427 @default.
• W2023062008 hasConceptScore W2023062008C153258448 @default.
• W2023062008 hasConceptScore W2023062008C154945302 @default.
• W2023062008 hasConceptScore W2023062008C158693339 @default.
• W2023062008 hasConceptScore W2023062008C186633575 @default.
• W2023062008 hasConceptScore W2023062008C203616005 @default.
• W2023062008 hasConceptScore W2023062008C28826006 @default.
• W2023062008 hasConceptScore W2023062008C31258907 @default.
• W2023062008 hasConceptScore W2023062008C33923547 @default.
• W2023062008 hasConceptScore W2023062008C41008148 @default.
• W2023062008 hasConceptScore W2023062008C42355184 @default.
• W2023062008 hasConceptScore W2023062008C42747912 @default.
• W2023062008 hasConceptScore W2023062008C50644808 @default.
• W2023062008 hasConceptScore W2023062008C57493831 @default.
• W2023062008 hasConceptScore W2023062008C62520636 @default.
• W2023062008 hasConceptScore W2023062008C81184566 @default.
• W2023062008 hasIssue "8" @default.

• next