Literature DB >> 12662686

Nonlinear independent component analysis: Existence and uniqueness results.

Aapo Hyvärinen1, Petteri Pajunen.   

Abstract

The question of existence and uniqueness of solutions for nonlinear independent component analysis is addressed. It is shown that if the space of mixing functions is not limited there exists always an infinity of solutions. In particular, it is shown how to construct parameterized families of solutions. The indeterminacies involved are not trivial, as in the linear case. Next, it is shown how to utilize some results of complex analysis to obtain uniqueness of solutions. We show that for two dimensions, the solution is unique up to a rotation, if the mixing function is constrained to be a conformal mapping together with some other assumptions. We also conjecture that the solution is strictly unique except in some degenerate cases, as the indeterminacy implied by the rotation is essentially similar to estimating the model of linear ICA.

Entities:  

Year:  1999        PMID: 12662686     DOI: 10.1016/s0893-6080(98)00140-3

Source DB:  PubMed          Journal:  Neural Netw        ISSN: 0893-6080


  21 in total

1.  How does spatial extent of fMRI datasets affect independent component analysis decomposition?

Authors:  Adriana Aragri; Tommaso Scarabino; Erich Seifritz; Silvia Comani; Sossio Cirillo; Gioacchino Tedeschi; Fabrizio Esposito; Francesco Di Salle
Journal:  Hum Brain Mapp       Date:  2006-09       Impact factor: 5.038

2.  Anatomical and functional assemblies of brain BOLD oscillations.

Authors:  Alexis T Baria; Marwan N Baliki; Todd Parrish; A Vania Apkarian
Journal:  J Neurosci       Date:  2011-05-25       Impact factor: 6.167

3.  Brain activity for spontaneous pain of postherpetic neuralgia and its modulation by lidocaine patch therapy.

Authors:  P Y Geha; M N Baliki; D R Chialvo; R N Harden; J A Paice; A V Apkarian
Journal:  Pain       Date:  2006-10-25       Impact factor: 6.961

4.  Low-Dimensional Density Ratio Estimation for Covariate Shift Correction.

Authors:  Petar Stojanov; Mingming Gong; Jaime G Carbonell; Kun Zhang
Journal:  Proc Mach Learn Res       Date:  2019-04

5.  Nonlinear extraction of independent components of natural images using radial gaussianization.

Authors:  Siwei Lyu; Eero P Simoncelli
Journal:  Neural Comput       Date:  2009-06       Impact factor: 2.026

6.  Deep Independence Network Analysis of Structural Brain Imaging: Application to Schizophrenia.

Authors:  Eduardo Castro; R Devon Hjelm; Sergey M Plis; Laurent Dinh; Jessica A Turner; Vince D Calhoun
Journal:  IEEE Trans Med Imaging       Date:  2016-02-11       Impact factor: 10.048

7.  Inference and Learning in a Latent Variable Model for Beta Distributed Interval Data.

Authors:  Hamid Mousavi; Mareike Buhl; Enrico Guiraud; Jakob Drefs; Jörg Lücke
Journal:  Entropy (Basel)       Date:  2021-04-29       Impact factor: 2.524

8.  Perfect Density Models Cannot Guarantee Anomaly Detection.

Authors:  Charline Le Lan; Laurent Dinh
Journal:  Entropy (Basel)       Date:  2021-12-16       Impact factor: 2.524

Review 9.  Ten Key Observations on the Analysis of Resting-state Functional MR Imaging Data Using Independent Component Analysis.

Authors:  Vince D Calhoun; Nina de Lacy
Journal:  Neuroimaging Clin N Am       Date:  2017-08-18       Impact factor: 2.264

10.  Nonlinear ICA of fMRI reveals primitive temporal structures linked to rest, task, and behavioral traits.

Authors:  Hiroshi Morioka; Vince Calhoun; Aapo Hyvärinen
Journal:  Neuroimage       Date:  2020-05-30       Impact factor: 6.556

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.