Matches in SemOpenAlex for { <https://semopenalex.org/work/W2155111216> ?p ?o ?g. }
- W2155111216 abstract "The variational Bayesian nonlinear blind source separation method introduced by Lappalainen and Honkela in 2000 is initialised with linear principal component analysis (PCA). Because of the multilayer perceptron (MLP) network used to model the nonlinearity, the method is susceptible to local minima and therefore sensitive to the initialisation used. As the method is used for nonlinear separation, the linear initialisation may in some cases lead it astray. In this paper we study the use of kernel PCA (KPCA) in the initialisation. KPCA is a rather straightforward generalisation of linear PCA and it is much faster to compute than the variational Bayesian method. The experiments show that it can produce significantly better initialisations than linear PCA. Additionally, the model comparison methods provided by the variational Bayesian framework can be easily applied to compare different kernels." @default.
- W2155111216 created "2016-06-24" @default.
- W2155111216 creator A5050971487 @default.
- W2155111216 creator A5062835183 @default.
- W2155111216 creator A5071329195 @default.
- W2155111216 creator A5082873543 @default.
- W2155111216 date "2004-01-01" @default.
- W2155111216 modified "2023-10-12" @default.
- W2155111216 title "Independent Component Analysis and Blind Signal Separation" @default.
- W2155111216 cites W1490760466 @default.
- W2155111216 cites W1528993561 @default.
- W2155111216 cites W1548802052 @default.
- W2155111216 cites W1562345118 @default.
- W2155111216 cites W1574202531 @default.
- W2155111216 cites W1586727544 @default.
- W2155111216 cites W1973737002 @default.
- W2155111216 cites W1983821361 @default.
- W2155111216 cites W2047229728 @default.
- W2155111216 cites W2084812512 @default.
- W2155111216 cites W2122664326 @default.
- W2155111216 cites W2140095548 @default.
- W2155111216 cites W2156436462 @default.
- W2155111216 cites W2799061466 @default.
- W2155111216 cites W2912934387 @default.
- W2155111216 cites W3085162807 @default.
- W2155111216 cites W33340904 @default.
- W2155111216 doi "https://doi.org/10.1007/b100528" @default.
- W2155111216 hasPublicationYear "2004" @default.
- W2155111216 type Work @default.
- W2155111216 sameAs 2155111216 @default.
- W2155111216 citedByCount "20" @default.
- W2155111216 countsByYear W21551112162012 @default.
- W2155111216 countsByYear W21551112162014 @default.
- W2155111216 countsByYear W21551112162015 @default.
- W2155111216 countsByYear W21551112162016 @default.
- W2155111216 countsByYear W21551112162017 @default.
- W2155111216 countsByYear W21551112162018 @default.
- W2155111216 countsByYear W21551112162023 @default.
- W2155111216 crossrefType "book" @default.
- W2155111216 hasAuthorship W2155111216A5050971487 @default.
- W2155111216 hasAuthorship W2155111216A5062835183 @default.
- W2155111216 hasAuthorship W2155111216A5071329195 @default.
- W2155111216 hasAuthorship W2155111216A5082873543 @default.
- W2155111216 hasBestOaLocation W21551112162 @default.
- W2155111216 hasConcept C104267543 @default.
- W2155111216 hasConcept C119857082 @default.
- W2155111216 hasConcept C120317606 @default.
- W2155111216 hasConcept C121332964 @default.
- W2155111216 hasConcept C127162648 @default.
- W2155111216 hasConcept C154945302 @default.
- W2155111216 hasConcept C168167062 @default.
- W2155111216 hasConcept C199360897 @default.
- W2155111216 hasConcept C2776061190 @default.
- W2155111216 hasConcept C2779843651 @default.
- W2155111216 hasConcept C28490314 @default.
- W2155111216 hasConcept C41008148 @default.
- W2155111216 hasConcept C51432778 @default.
- W2155111216 hasConcept C76155785 @default.
- W2155111216 hasConcept C84462506 @default.
- W2155111216 hasConcept C9390403 @default.
- W2155111216 hasConcept C97355855 @default.
- W2155111216 hasConceptScore W2155111216C104267543 @default.
- W2155111216 hasConceptScore W2155111216C119857082 @default.
- W2155111216 hasConceptScore W2155111216C120317606 @default.
- W2155111216 hasConceptScore W2155111216C121332964 @default.
- W2155111216 hasConceptScore W2155111216C127162648 @default.
- W2155111216 hasConceptScore W2155111216C154945302 @default.
- W2155111216 hasConceptScore W2155111216C168167062 @default.
- W2155111216 hasConceptScore W2155111216C199360897 @default.
- W2155111216 hasConceptScore W2155111216C2776061190 @default.
- W2155111216 hasConceptScore W2155111216C2779843651 @default.
- W2155111216 hasConceptScore W2155111216C28490314 @default.
- W2155111216 hasConceptScore W2155111216C41008148 @default.
- W2155111216 hasConceptScore W2155111216C51432778 @default.
- W2155111216 hasConceptScore W2155111216C76155785 @default.
- W2155111216 hasConceptScore W2155111216C84462506 @default.
- W2155111216 hasConceptScore W2155111216C9390403 @default.
- W2155111216 hasConceptScore W2155111216C97355855 @default.
- W2155111216 hasLocation W21551112161 @default.
- W2155111216 hasLocation W21551112162 @default.
- W2155111216 hasLocation W21551112163 @default.
- W2155111216 hasLocation W21551112164 @default.
- W2155111216 hasLocation W21551112165 @default.
- W2155111216 hasLocation W21551112166 @default.
- W2155111216 hasLocation W21551112167 @default.
- W2155111216 hasOpenAccess W2155111216 @default.
- W2155111216 hasPrimaryLocation W21551112161 @default.
- W2155111216 hasRelatedWork W2046761971 @default.
- W2155111216 hasRelatedWork W2103029460 @default.
- W2155111216 hasRelatedWork W2156932837 @default.
- W2155111216 hasRelatedWork W2361066326 @default.
- W2155111216 hasRelatedWork W2362299104 @default.
- W2155111216 hasRelatedWork W2364896863 @default.
- W2155111216 hasRelatedWork W2380698615 @default.
- W2155111216 hasRelatedWork W2390344110 @default.
- W2155111216 hasRelatedWork W374502268 @default.
- W2155111216 hasRelatedWork W2182042810 @default.
- W2155111216 isParatext "false" @default.
- W2155111216 isRetracted "false" @default.
- W2155111216 magId "2155111216" @default.