Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Topographic independent component analysis.

A Hyvärinen, P O Hoyer, M Inki

    Neural Computation
    |July 7, 2001
    PubMed
    Summary

    This study introduces a novel topographic ordering for independent component analysis (ICA) components. By leveraging residual dependencies, it creates a meaningful spatial representation of otherwise independent signals.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    Aerosol generation during coughing: an observational study.

    The Journal of laryngology and otology·2022
    Same author

    Household dampness and microbial exposure related to allergy and respiratory health in Danish adults.

    European clinical respiratory journal·2020
    Same author

    Decoding attentional states for neurofeedback: Mindfulness vs. wandering thoughts.

    NeuroImage·2018
    Same author

    Nasal mucosa and blood cell transcriptome profiles do not reflect respiratory symptoms associated with moisture damage.

    Indoor air·2018
    Same author

    Early age exposure to moisture damage and systemic inflammation at the age of 6 years.

    Indoor air·2018
    Same author

    Microbial growth in building material samples and occupants' health in severely moisture-damaged homes.

    Indoor air·2017

    Area of Science:

    • Computational Neuroscience
    • Signal Processing
    • Machine Learning

    Background:

    • Standard independent component analysis (ICA) assumes component independence, often not met in practice.
    • Residual dependencies between estimated components are typically ignored.
    • Lack of inherent order in traditional ICA components limits interpretability.

    Purpose of the Study:

    • To propose a method for defining a topographic order for ICA components.
    • To utilize residual dependence structures for component organization.
    • To create a spatially meaningful representation of approximately independent components.

    Main Methods:

    • Defined a distance metric between components based on higher-order correlations.
    • Employed this distance to construct a topographic representation of components.
    • Utilized linear decomposition to obtain approximately independent components.

    Main Results:

    • Successfully developed a topographic representation for ICA components.
    • Demonstrated that component proximity in the topography approximates their dependence.
    • Showcased a method to order components based on their residual correlations.

    Conclusions:

    • Residual dependencies in ICA can be leveraged to establish a meaningful topographic order.
    • The proposed topographic representation enhances the interpretability of ICA results.
    • This approach provides a structured decomposition into approximately independent and ordered components.

    Related Experiment Videos