000861439 001__ 861439
000861439 005__ 20210130000818.0
000861439 0247_ $$2Handle$$a2128/21854
000861439 037__ $$aFZJ-2019-01911
000861439 1001_ $$0P:(DE-Juel1)173704$$aMuzzarelli, Laura$$b0$$eCorresponding author$$ufzj
000861439 1112_ $$a2019 International Joint Conference on Neural Networks$$cBudapest$$d2019-07-14 - 2019-07-19$$wHungary
000861439 245__ $$aRank Selection in Non-negative Matrix Factorization: systematic comparison and a new MAD metric
000861439 260__ $$c2019
000861439 300__ $$a7
000861439 3367_ $$2ORCID$$aCONFERENCE_PAPER
000861439 3367_ $$033$$2EndNote$$aConference Paper
000861439 3367_ $$2BibTeX$$aINPROCEEDINGS
000861439 3367_ $$2DRIVER$$aconferenceObject
000861439 3367_ $$2DataCite$$aOutput Types/Conference Paper
000861439 3367_ $$0PUB:(DE-HGF)8$$2PUB:(DE-HGF)$$aContribution to a conference proceedings$$bcontrib$$mcontrib$$s1552655642_21952
000861439 500__ $$aThis study was partly supported by the Helmholtz Portfolio Theme "Supercomputing and Modeling for the Human Brain" and the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 785907 (HBP SGA2).
000861439 520__ $$aAbstract—Non-Negative Matrix Factorization (NMF) is a powerful dimensionality reduction and factorization method that provides a part-based representation of the data. In the absence of a priori knowledge about the latent dimensionality of the data, it is necessary to select a rank of the reduced representation. Several rank selection methods have been proposed, but no consensus exists on when a method is suitable to use. In this work, we propose a new metric for rank selection based on imputation cross-validation, and we systematically compare it against six other metrics while assessing the effects of data properties. Using synthetic datasets with different properties, our work critically evidences that most methods fail to identify the true rank. We show that properties of the data heavily impact the ability of different methods. Imputation-based metrics, including our new MADimput, provided the best accuracy irrespective of the data type, but no solution worked perfectly in all circumstances. One should therefore carefully assess characteristics of their dataset in order to identify the most suitable metric for rank selection. Keywords— non-negative matrix factorization, rank selection, cross-validation.
000861439 536__ $$0G:(DE-HGF)POF3-574$$a574 - Theory, modelling and simulation (POF3-574)$$cPOF3-574$$fPOF III$$x0
000861439 536__ $$0G:(DE-Juel1)HGF-SMHB-2013-2017$$aSMHB - Supercomputing and Modelling for the Human Brain (HGF-SMHB-2013-2017)$$cHGF-SMHB-2013-2017$$fSMHB$$x1
000861439 536__ $$0G:(EU-Grant)785907$$aHBP SGA2 - Human Brain Project Specific Grant Agreement 2 (785907)$$c785907$$fH2020-SGA-FETFLAG-HBP-2017$$x2
000861439 7001_ $$0P:(DE-Juel1)172811$$aWeis, Susanne$$b1$$ufzj
000861439 7001_ $$0P:(DE-Juel1)131678$$aEickhoff, Simon$$b2$$ufzj
000861439 7001_ $$0P:(DE-Juel1)172843$$aPatil, Kaustubh$$b3$$ufzj
000861439 8564_ $$uhttps://juser.fz-juelich.de/record/861439/files/NMFrankselection_20181214m10_submit_withTitleAuthAbs19.pdf$$yOpenAccess
000861439 8564_ $$uhttps://juser.fz-juelich.de/record/861439/files/NMFrankselection_20181214m10_submit_withTitleAuthAbs19.pdf?subformat=pdfa$$xpdfa$$yOpenAccess
000861439 909CO $$ooai:juser.fz-juelich.de:861439$$pdnbdelivery$$pec_fundedresources$$pVDB$$pdriver$$popen_access$$popenaire
000861439 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)173704$$aForschungszentrum Jülich$$b0$$kFZJ
000861439 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172811$$aForschungszentrum Jülich$$b1$$kFZJ
000861439 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131678$$aForschungszentrum Jülich$$b2$$kFZJ
000861439 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)172843$$aForschungszentrum Jülich$$b3$$kFZJ
000861439 9131_ $$0G:(DE-HGF)POF3-574$$1G:(DE-HGF)POF3-570$$2G:(DE-HGF)POF3-500$$3G:(DE-HGF)POF3$$4G:(DE-HGF)POF$$aDE-HGF$$bKey Technologies$$lDecoding the Human Brain$$vTheory, modelling and simulation$$x0
000861439 9141_ $$y2019
000861439 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
000861439 920__ $$lyes
000861439 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
000861439 980__ $$acontrib
000861439 980__ $$aVDB
000861439 980__ $$aUNRESTRICTED
000861439 980__ $$aI:(DE-Juel1)INM-7-20090406
000861439 9801_ $$aFullTexts