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

On-line learning through simple perceptron learning with a margin.

Kazuyuki Hara1, Masato Okada

  • 1Department of Electronics and Information Engineering, Tokyo Metropolitan College of Technology, 1-10-40 Higashi-oi, Shinagawa-ku, Tokyo 140-0011, Japan. hara@tokyo-tmct.ac.jp

Neural Networks : the Official Journal of the International Neural Network Society
|March 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

Comparison of the PK, PD, safety, tolerability, and immunogenicity of proposed biosimilar RGB-19 and tocilizumab in healthy Japanese males: a phase 1, randomised, crossover study.

EULAR rheumatology open·2026
Same author

Efficacy, safety, and immunogenicity of proposed biosimilar RGB-19 and tocilizumab intravenously administered to adults with active rheumatoid arthritis and an inadequate response to methotrexate: a phase 3, randomised study.

EULAR rheumatology open·2026
Same author

Impact of Diabetes on the Efficacy of Low-Voltage Area Ablation for Persistent Atrial Fibrillation: A Subanalysis of the SUPPRESS-AF Trial.

Journal of the American Heart Association·2026
Same author

Efficacy of Low-Voltage Area Ablation Across Substrate Size in Persistent Atrial Fibrillation: A Post Hoc Analysis of the SUPPRESS-AF Randomized Trial.

Journal of the American Heart Association·2026
Same author

Induction indices during ventricular pacing as an alternative diagnostic tool in supraventricular tachycardias.

Heart rhythm O2·2026
Same author

Triglyceride-Glucose Index and Clinical Outcome in Patients With Heart Failure With Preserved Ejection Fraction.

JACC. Asia·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
Same journal

IGFD-Net: Illumination-guided frequency decoupling for polarization image fusion.

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

Multiple-Strategies dung beetle optimizer and its applications in engineering optimization and bankruptcy prediction.

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

Aggregating global-scale pixel-wise forgery cues within a graph.

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

Finite-Time intermittent control for secure synchronization of Neutral-Type stochastic delayed neural networks under aperiodic DoS attacks.

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

This study introduces a novel learning method with a margin kappa, outperforming standard perceptron and Hebbian learning in early stages. Its long-term learning curve mirrors perceptron learning, with adaptive margin control also explored.

Area of Science:

  • Machine Learning
  • Computational Neuroscience
  • Artificial Intelligence

Background:

  • Perceptron learning and Hebbian learning are foundational algorithms in neural networks.
  • Understanding generalization ability is crucial for effective machine learning models.
  • The role of margins in learning algorithms can significantly impact performance.

Purpose of the Study:

  • To analyze a learning method incorporating a margin kappa, inspired by Gardner's work.
  • To compare the generalization ability of this new method against perceptron and Hebbian learning.
  • To investigate the asymptotic properties and adaptive control of the margin-based learning method.

Main Methods:

  • Analysis of a perceptron learning variant with a margin parameter kappa.

Related Experiment Videos

  • Comparative study of generalization performance at early learning stages.
  • Computer simulations to examine asymptotic learning curve properties.
  • Investigation into adaptive margin control strategies.

    • Main Results:

    • The margin kappa learning method demonstrated superior generalization ability compared to perceptron and Hebbian learning in the initial learning phase.
    • The asymptotic learning curve of the margin kappa method was found to be identical to that of standard perceptron learning.
    • An adaptive margin control method was also explored.

    Conclusions:

    • The margin kappa learning method offers an advantage in early-stage generalization for simple perceptron tasks.
    • The method's long-term behavior converges with traditional perceptron learning.
    • Adaptive margin control presents a potential avenue for further optimization.