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

Radial Basis Function Network Configuration Using Mutual Information and the Orthogonal Least Squares Algorithm.

Steve A. Billings1, Guang L. Zheng

  • 1University of Sheffield, UK

Neural Networks : the Official Journal of the International Neural Network Society
|December 1, 1996
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

On-line identification of nonlinear systems using Volterra polynomial basis function neural networks.

Neural networks : the official journal of the International Neural Network Society·2003
Same author

Dual-orthogonal radial basis function networks for nonlinear time series prediction.

Neural networks : the official journal of the International Neural Network Society·2003
Same author

On-line Supervised Adaptive Training Using Radial Basis Function Networks.

Neural networks : the official journal of the International Neural Network Society·1996
Same journal

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Physics-encoded convolutional neural operators for parametric PDEs: A convergence-guaranteed framework via pre-computed kernel fields.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Exploiting audio-visual modalities in videos: Object detection via multi-stage bilateral coupling network.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Reliability-aware modality completion with cross-modal distillation for federated learning with missing modalities.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a novel method for selecting optimal input variables for neural networks using mutual information. This approach enhances network performance by identifying relevant and non-redundant input features, reducing computational complexity.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Information Theory

Background:

  • Neural network input node selection often relies on prior knowledge or trial-and-error, which can lead to suboptimal performance.
  • Irrelevant or redundant input variables can degrade neural network accuracy and increase computational costs.

Purpose of the Study:

  • To develop a heuristic algorithm for selecting a suboptimal set of input variables for neural networks.
  • To improve neural network performance by utilizing the information content relevant to the network's output.

Main Methods:

  • Utilized mutual information to quantify the relationship between input variables and network output.
  • Selected input variables based on high mutual information with the output and low dependence on other selected variables.

Related Experiment Videos

  • Employed radial basis function (RBF) networks trained with the orthogonal least squares (OLS) algorithm for performance assessment.

    • Main Results:

    • Demonstrated the effectiveness of the proposed input variable selection algorithms on both real and simulated datasets.
    • The method successfully identified relevant and non-redundant input features, leading to improved network efficiency.
    • Orthogonal least squares algorithm effectively selected hidden layer nodes based on error reduction ratios.

    Conclusions:

    • The proposed mutual information-based approach offers an effective strategy for input variable selection in neural networks.
    • This method enhances network performance by optimizing input feature selection, reducing complexity and computational load.
    • The integration with RBF networks and OLS training provides a robust framework for practical applications.