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

Generalized hamming networks and applications.

Konstantinos Koutroumbas1, Nicholas Kalouptsidis

  • 1Institute for Space Applications and Remote Sensing, National Observatory of Athens, Metaxa and V. Pavlou, Palaia Penteli, 15236, Athens, Greece. koutroum@space.noa.gr

Neural Networks : the Official Journal of the International Neural Network Society
|June 14, 2005
PubMed
Summary

This study generalizes the Hamming network, introducing a flexible model with time-varying dynamics and parameters. The enhanced Hamming maxnet offers improved convergence analysis and stabilization, demonstrating practical advantages in hardware implementations.

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

Discriminating benign from malignant thyroid lesions using artificial intelligence and statistical selection of morphometric features.

Oncology reports·2006
Same author

The potential of feature selection by statistical techniques and the use of statistical classifiers in the discrimination of benign from malignant gastric lesions.

Oncology reports·2006
See all related articles

Area of Science:

  • Artificial Intelligence
  • Computational Neuroscience
  • Machine Learning

Background:

  • The classical Hamming network is a foundational model in neural computation.
  • Existing versions of the Hamming Maxnet have limitations in flexibility and parameter adaptability.
  • Time-varying dynamics and generalized non-linear functions are not fully explored in standard Hamming networks.

Purpose of the Study:

  • To generalize the classical Hamming network, specifically the Hamming maxnet, by introducing time-varying dynamics and parameters.
  • To provide a detailed convergence analysis for the generalized model, including bounds on iterations and distribution functions.
  • To explore stabilization mechanisms and compare the generalized scheme with existing Hamming network variants in terms of convergence time and hardware implementation.

Related Experiment Videos

Main Methods:

  • Development of a generalized Hamming maxnet model with time-varying non-linear functions and weights.
  • Mathematical derivation of convergence bounds and analysis of iteration distribution functions for uniform and peak initial distributions.
  • Implementation of stabilization mechanisms to prevent divergence or decay of node values.
  • Simulation-based comparison of the generalized model against original Hamming maxnet and its variants.

Main Results:

  • The proposed generalized Hamming maxnet model accommodates a wider range of existing versions.
  • A detailed convergence analysis is provided, with derived bounds on iterations and distribution functions.
  • Stabilization mechanisms are described and shown to be effective.
  • Simulations demonstrate the advantages of the generalized extension, including improved convergence time in hardware implementations.

Conclusions:

  • The generalized Hamming network offers enhanced flexibility and performance over classical models.
  • The detailed convergence analysis and stabilization mechanisms contribute to a more robust computational model.
  • The generalized Hamming network framework has potential applications in classification, clustering, vector quantization, and function optimization.