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

Asymmetric kernel regression.

Mark Mackenzie1, A Kiet Tieu

  • 1Mechanical Engineering, University of Wollongong, Wollongong, N.S.W. 2522, Australia.

IEEE Transactions on Neural Networks
|September 24, 2004
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

Evaluating the Impact of a System-Wide Prehospital Teleconsultation on Rural Stroke Transportation Efficiency.

Stroke (Hoboken, N.J.)·2026
Same author

A Molecular Dynamics Study of Boundary Lubrication of Silicate Glass at High Temperatures.

Langmuir : the ACS journal of surfaces and colloids·2025
Same author

A Randomized Crossover Simulation Study of Prolonged Conventional Versus Modified "Koch" Chest Compressions in a Height-Restricted Setting.

Air medical journal·2025
Same author

Caring for the invisible and forgotten: a qualitative document analysis and experience-based co-design project to improve the care of families experiencing out-of-hospital cardiac arrest.

CJEM·2023
Same author

Near-zero-index ultra-fast pulse characterization.

Nature communications·2022
Same author

First-principles molecular dynamics study on the surface chemistry and nanotribological properties of MgAl layered double hydroxides.

Nanoscale·2021
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

Boundary errors in kernel regression can reduce accuracy. This study introduces asymmetric kernel functions to correct these errors near data boundaries, improving radial-basis neural network design without compromising noise filtering.

Area of Science:

  • Machine Learning
  • Artificial Intelligence
  • Data Science

Background:

  • Kernel regression is a key model for explaining and designing radial-basis neural networks.
  • Practical applications reveal significant bias errors at data boundaries, impacting regression accuracy.
  • Existing symmetric kernel functions struggle with boundary effects.

Purpose of the Study:

  • To investigate and correct boundary bias errors in kernel regression.
  • To enhance the accuracy of kernel regression, particularly for radial-basis neural networks.
  • To introduce a novel approach using asymmetric kernel functions.

Main Methods:

  • Substitution of asymmetric kernel functions for symmetric ones at boundary data points.
  • Analysis of kernel regression performance with the new asymmetric kernel approach.

Related Experiment Videos

  • Evaluation of noise-filtering properties alongside boundary error correction.

    • Main Results:

    • The asymmetric kernel function effectively corrects boundary bias errors.
    • Improved accuracy in kernel regression models near data boundaries was achieved.
    • Noise-filtering capabilities of the kernel regression were maintained.

    Conclusions:

    • Asymmetric kernel functions offer a viable solution for boundary errors in kernel regression.
    • This method enhances the practical application of kernel regression in designing neural networks.
    • The approach allows for closer boundary approximation without sacrificing data smoothing.