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

Focused local learning with wavelet neural networks.

E A Rying1, G L Bilbro, Jye-Chyi Lu

  • 1Dept. of Electr. and Comput. Eng., North Carolina State Univ., Raleigh, NC.

IEEE Transactions on Neural Networks
|February 5, 2008
PubMed
Summary
This summary is machine-generated.

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

A risk-sharing-based resilient renewable energy supply network model under the COVID-19 pandemic.

Sustainable production and consumption·2023
Same author

Maximum likelihood estimation of time to first event in the presence of data gaps and multiple events.

Statistical methods in medical research·2012
Same author

Extending the skill test for disease diagnosis.

Statistics in medicine·2008
Same author

A vertical-energy-thresholding procedure for data reduction with multiple complex curves.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2006
Same author

Left-ventricle boundary detection from nuclear medicine images.

Journal of digital imaging·1998
Same author

Noise removal from multiple MRI images.

Journal of digital imaging·1994
Same journal

Universal perceptron and DNA-like learning algorithm for binary neural networks: LSBF and PBF implementations.

IEEE transactions on neural networks·2013
Same journal

Guest editorial: special section on white box nonlinear prediction models.

IEEE transactions on neural networks·2011
Same journal

Data-based fault-tolerant control of high-speed trains with traction/braking notch nonlinearities and actuator failures.

IEEE transactions on neural networks·2011
Same journal

Guest editorial: special section on data-based control, modeling, and optimization.

IEEE transactions on neural networks·2011
Same journal

Neural network-based multiple robot simultaneous localization and mapping.

IEEE transactions on neural networks·2011
Same journal

Data-driven model-free adaptive control for a class of MIMO nonlinear discrete-time systems.

IEEE transactions on neural networks·2011
See all related articles

This study introduces a new objective function for wavelet neural networks, improving model accuracy by minimizing both local and global errors. The methods enhance wavelet selection for better performance with minimal computational cost.

Area of Science:

  • Artificial Intelligence
  • Signal Processing
  • Machine Learning

Background:

  • Wavelet neural networks (WNNs) are powerful tools for signal processing and machine learning.
  • Existing WNN construction methods often struggle to balance global accuracy with the precise modeling of local features.
  • Model parsimony is crucial for efficient and interpretable WNNs.

Purpose of the Study:

  • To present a novel objective function for WNN construction that integrates local errors, global errors, and model parsimony.
  • To introduce two new methodologies for minimizing this objective function, with a specific focus on reducing local error.
  • To demonstrate the effectiveness of the proposed methods in improving WNN performance and efficiency.

Main Methods:

  • Development of a novel objective function incorporating local/global errors and model parsimony.

Related Experiment Videos

  • Utilizing a locally adaptive grid during network initialization to select relevant wavelet basis functions.
  • Employing a subspace projection operator for focused wavelet basis function selection during network construction.

    • Main Results:

    • The proposed methods effectively minimize both local and global errors in WNNs.
    • Model parsimony is maintained, preventing overfitting and enhancing interpretability.
    • The computational complexity increase is minimal, ensuring practical applicability.

    Conclusions:

    • The novel objective function and associated methods represent a significant advancement in WNN construction.
    • These techniques enable the development of more accurate and efficient WNNs for complex data analysis.
    • The approach offers a robust solution for addressing local features while optimizing global performance.