Matches in SemOpenAlex for { <https://semopenalex.org/work/W2001096515> ?p ?o ?g. }
- W2001096515 endingPage "698" @default.
- W2001096515 startingPage "677" @default.
- W2001096515 abstract "Nonnegative matrix factorization (NMF) under the separability assumption can provably be solved efficiently, even in the presence of noise, and has been shown to be a powerful technique in document classification and hyperspectral unmixing. This problem is referred to as near-separable NMF and requires that there exists a cone spanned by a small subset of the columns of the input nonnegative matrix approximately containing all columns. In this paper, we propose a preconditioning based on semidefinite programming making the input matrix well-conditioned. This in turn can improve significantly the performance of near-separable NMF algorithms which is illustrated on the popular successive projection algorithm (SPA). The new preconditioned SPA is provably more robust to noise, and outperforms SPA on several synthetic data sets. We also show how an active-set method allow us to apply the preconditioning on large-scale real-world hyperspectral images." @default.
- W2001096515 created "2016-06-24" @default.
- W2001096515 creator A5018511273 @default.
- W2001096515 creator A5040368041 @default.
- W2001096515 date "2015-01-01" @default.
- W2001096515 modified "2023-09-25" @default.
- W2001096515 title "Semidefinite Programming Based Preconditioning for More Robust Near-Separable Nonnegative Matrix Factorization" @default.
- W2001096515 cites W1522222833 @default.
- W2001096515 cites W1902027874 @default.
- W2001096515 cites W1976391658 @default.
- W2001096515 cites W1982755765 @default.
- W2001096515 cites W2039806043 @default.
- W2001096515 cites W2055356142 @default.
- W2001096515 cites W2056124912 @default.
- W2001096515 cites W2063790512 @default.
- W2001096515 cites W2064380440 @default.
- W2001096515 cites W2115163184 @default.
- W2001096515 cites W2118381521 @default.
- W2001096515 cites W2124172487 @default.
- W2001096515 cites W2125118959 @default.
- W2001096515 cites W2133629339 @default.
- W2001096515 cites W2138663260 @default.
- W2001096515 cites W2157321686 @default.
- W2001096515 cites W2158713176 @default.
- W2001096515 cites W3106158395 @default.
- W2001096515 doi "https://doi.org/10.1137/130940670" @default.
- W2001096515 hasPublicationYear "2015" @default.
- W2001096515 type Work @default.
- W2001096515 sameAs 2001096515 @default.
- W2001096515 citedByCount "26" @default.
- W2001096515 countsByYear W20010965152014 @default.
- W2001096515 countsByYear W20010965152015 @default.
- W2001096515 countsByYear W20010965152016 @default.
- W2001096515 countsByYear W20010965152017 @default.
- W2001096515 countsByYear W20010965152018 @default.
- W2001096515 countsByYear W20010965152019 @default.
- W2001096515 countsByYear W20010965152020 @default.
- W2001096515 countsByYear W20010965152021 @default.
- W2001096515 countsByYear W20010965152022 @default.
- W2001096515 countsByYear W20010965152023 @default.
- W2001096515 crossrefType "journal-article" @default.
- W2001096515 hasAuthorship W2001096515A5018511273 @default.
- W2001096515 hasAuthorship W2001096515A5040368041 @default.
- W2001096515 hasBestOaLocation W20010965152 @default.
- W2001096515 hasConcept C101901036 @default.
- W2001096515 hasConcept C106487976 @default.
- W2001096515 hasConcept C11413529 @default.
- W2001096515 hasConcept C121332964 @default.
- W2001096515 hasConcept C126255220 @default.
- W2001096515 hasConcept C134306372 @default.
- W2001096515 hasConcept C152671427 @default.
- W2001096515 hasConcept C153180895 @default.
- W2001096515 hasConcept C154945302 @default.
- W2001096515 hasConcept C158693339 @default.
- W2001096515 hasConcept C159078339 @default.
- W2001096515 hasConcept C159985019 @default.
- W2001096515 hasConcept C192562407 @default.
- W2001096515 hasConcept C33923547 @default.
- W2001096515 hasConcept C41008148 @default.
- W2001096515 hasConcept C42355184 @default.
- W2001096515 hasConcept C62520636 @default.
- W2001096515 hasConcept C70710897 @default.
- W2001096515 hasConceptScore W2001096515C101901036 @default.
- W2001096515 hasConceptScore W2001096515C106487976 @default.
- W2001096515 hasConceptScore W2001096515C11413529 @default.
- W2001096515 hasConceptScore W2001096515C121332964 @default.
- W2001096515 hasConceptScore W2001096515C126255220 @default.
- W2001096515 hasConceptScore W2001096515C134306372 @default.
- W2001096515 hasConceptScore W2001096515C152671427 @default.
- W2001096515 hasConceptScore W2001096515C153180895 @default.
- W2001096515 hasConceptScore W2001096515C154945302 @default.
- W2001096515 hasConceptScore W2001096515C158693339 @default.
- W2001096515 hasConceptScore W2001096515C159078339 @default.
- W2001096515 hasConceptScore W2001096515C159985019 @default.
- W2001096515 hasConceptScore W2001096515C192562407 @default.
- W2001096515 hasConceptScore W2001096515C33923547 @default.
- W2001096515 hasConceptScore W2001096515C41008148 @default.
- W2001096515 hasConceptScore W2001096515C42355184 @default.
- W2001096515 hasConceptScore W2001096515C62520636 @default.
- W2001096515 hasConceptScore W2001096515C70710897 @default.
- W2001096515 hasIssue "1" @default.
- W2001096515 hasLocation W20010965151 @default.
- W2001096515 hasLocation W20010965152 @default.
- W2001096515 hasOpenAccess W2001096515 @default.
- W2001096515 hasPrimaryLocation W20010965151 @default.
- W2001096515 hasRelatedWork W2020769413 @default.
- W2001096515 hasRelatedWork W2028051535 @default.
- W2001096515 hasRelatedWork W2338894643 @default.
- W2001096515 hasRelatedWork W2374719179 @default.
- W2001096515 hasRelatedWork W2501002973 @default.
- W2001096515 hasRelatedWork W2513087981 @default.
- W2001096515 hasRelatedWork W2776964412 @default.
- W2001096515 hasRelatedWork W2898368675 @default.
- W2001096515 hasRelatedWork W2972457993 @default.
- W2001096515 hasRelatedWork W2161790104 @default.
- W2001096515 hasVolume "25" @default.
- W2001096515 isParatext "false" @default.
- W2001096515 isRetracted "false" @default.