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 competitive learning.

L Wang1

  • 1Sch. of Comput. and Math., Deakin Univ., Geelong, Vic.

IEEE Transactions on Neural Networks
|January 1, 1997
PubMed
Summary
This summary is machine-generated.

This study introduces learning rates ensuring equal statistical importance for all training patterns, making outcomes independent of presentation order. Weight normalization methods yield similar results when competitive neurons consistently win the same patterns.

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

Prodomain mutations at the subtilisin interface: correlation of binding energy and the rate of catalyzed folding.

Biochemistry·1995
Same author

Synthesis and characterization of a highly potent and effective antagonist of retinoic acid receptors.

Journal of medicinal chemistry·1995
Same author

Thermodynamic parameters for the helix-coil transition of oligopeptides: molecular dynamics simulation with the peptide growth method.

Proceedings of the National Academy of Sciences of the United States of America·1995
Same author

The sixty nucleotide OccR operator contains a subsite essential and sufficient for OccR binding and a second subsite required for ligand-responsive DNA bending.

Journal of molecular biology·1995
Same author

Cloning of a cDNA encoding a 190-kDa insulin receptor substrate-1-like protein of simian COS cells.

Biochemical and biophysical research communications·1995
Same author

Differences in oxidative response of subpopulations of neutrophils from healthy subjects and patients with rheumatoid arthritis.

Annals of the rheumatic diseases·1995
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

Area of Science:

  • Computational Neuroscience
  • Machine Learning

Background:

  • Order-dependent learning can bias neural network training.
  • Weight normalization techniques aim to stabilize learning.

Purpose of the Study:

  • Derive learning rates for order-independent training.
  • Compare learning rules under different weight normalization constraints.

Main Methods:

  • Statistical derivation of learning rates.
  • Analysis of competitive neuron behavior.
  • Computer simulations to validate theoretical results.

Main Results:

  • Developed learning rates ensure equal statistical importance of training patterns.
  • Learning outcomes are independent of training pattern presentation order.

Related Experiment Videos

  • Length-constraint and sum-constraint weight normalization yield comparable results.

    • Conclusions:

    • Proposed learning rates enhance the robustness of neural network training.
    • Order-independent learning is achievable with appropriate learning rate derivation.
    • Weight normalization methods converge to similar outcomes under specific conditions.