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

Fast neural network ensemble learning via negative-correlation data correction.

Zeke S H Chan, Nik Kasabov

    IEEE Transactions on Neural Networks
    |December 14, 2005
    PubMed
    Summary

    A novel negative correlation (NC) learning method is introduced, offering easier implementation and reduced communication overhead compared to standard approaches. This new NC method also supports ensembles of diverse neural networks.

    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

    Bayesian learning of sparse gene regulatory networks.

    Bio Systems·2007
    Same author

    A hybrid genetic algorithm and expectation maximization method for global gene trajectory clustering.

    Journal of bioinformatics and computational biology·2005
    See all related articles

    Area of Science:

    • Machine Learning
    • Artificial Intelligence
    • Neural Networks

    Background:

    • Standard negative correlation (NC) learning methods often involve significant communication overhead.
    • Applying existing NC methods to ensembles of heterogeneous networks can be challenging.

    Purpose of the Study:

    • To propose a new, easily implementable negative correlation (NC) learning method.
    • To reduce communication overhead in NC learning.
    • To enhance the applicability of NC learning to heterogeneous network ensembles.

    Main Methods:

    • Development of a novel negative correlation (NC) learning algorithm.
    • Implementation of the proposed NC method for comparative analysis.
    • Evaluation of communication overhead and network applicability.

    Related Experiment Videos

    Main Results:

    • The proposed NC learning method demonstrates significantly reduced communication overhead compared to standard NC methods.
    • The new NC method is successfully applied to ensembles of heterogeneous networks.
    • Ease of implementation is a key advantage of the proposed method.

    Conclusions:

    • The new negative correlation (NC) learning method offers a practical and efficient alternative.
    • This advancement facilitates the use of NC learning in more complex and diverse network architectures.
    • The method addresses key limitations of existing NC learning techniques.