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

K-winner networks.

W J Wolfe1, D Mathis, C Anderson

  • 1Dept. of Electr. Eng. and Comput. Sci., Colorado Univ., Denver, CO.

IEEE Transactions on Neural Networks
|January 1, 1991
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 sputum induction with fiber-optic bronchoscopy in the diagnosis of tuberculosis.

American journal of respiratory and critical care medicine·1995
Same author

Metabolic requirements for induction of contact hypersensitivity to immunotoxic polyaromatic hydrocarbons.

Journal of immunology (Baltimore, Md. : 1950)·1995
Same author

Nerves containing nitric oxide synthase and their possible function in the control of catecholamine secretion in the bovine adrenal medulla.

Journal of the autonomic nervous system·1995
Same author

Echocardiographic-morphologic correlations in tricuspid atresia.

Journal of the American College of Cardiology·1995
Same author

Drawing serum haptoglobin levels to determine hemolysis due to use of an incorrect reprocessed dialyzer.

ANNA journal·1995
Same author

Diode laser treatment for retinopathy of prematurity: structural and functional outcome.

The British journal of ophthalmology·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

This study presents parameters for reliable K-winner performance in mutually inhibitory networks. It details how initial activations and external inputs influence network convergence and identifies conditions for accurate K-winner selection.

Area of Science:

  • Computational Neuroscience
  • Artificial Neural Networks
  • Network Dynamics

Background:

  • Mutually inhibitory networks are crucial for decision-making and pattern recognition.
  • Understanding parameters for reliable K-winner performance is essential for network stability.
  • Existing models like the sigmoid model provide a baseline for comparison.

Purpose of the Study:

  • To analyze a special class of mutually inhibitory networks.
  • To present parameters for reliable K-winner performance.
  • To compare network dynamics with the sigmoid model.

Main Methods:

  • Modeling network dynamics using interactive activation.
  • Deriving network parameters based on initial activations and external inputs.

Related Experiment Videos

  • Analyzing anomalous behavior arising from mixed initial activations and external inputs.

    • Main Results:

    • Parameters for selecting units with larger initial activations (converging to the nearest stable state) were derived for equal external inputs.
    • Parameters for selecting units with larger external inputs (converging to the lowest energy stable state) were derived for equal initial activations.
    • Restrictions on initial states were identified to ensure accurate K-winner performance with unequal external inputs.

    Conclusions:

    • Network behavior is highly sensitive to the interplay between initial activations and external inputs.
    • Specific conditions can ensure reliable K-winner performance even with unequal inputs.
    • The findings contribute to the design and understanding of stable neural network architectures.