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

A learning algorithm for applying synthesized stable dynamics to system identification.

Gregory L. Heileman1, Chaouki T. Abdallah, James W. Howse

  • 1Department of Electrical and Computer Engineering, University of New Mexico, Albuquerque, USA

Neural Networks : the Official Journal of the International Neural Network Society
|March 29, 2003
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

Order of Search in Fuzzy ART and Fuzzy ARTMAP: Effect of the Choice Parameter.

Neural networks : the official journal of the International Neural Network Societyยท1996
See all related articles

This study introduces an input term to Cohen

Area of Science:

  • Dynamical Systems and Control Theory
  • Computational Neuroscience
  • Machine Learning

Background:

  • Cohen's models provide a framework for analyzing dynamical systems.
  • System identification requires models that can incorporate external influences.
  • Qualitative information, like attractor dimension, is challenging to encode directly into existing models.

Purpose of the Study:

  • To extend Cohen's models by incorporating an input term for enhanced system identification.
  • To enable the direct encoding of qualitative system properties into dynamical models.
  • To develop a stable and convergent learning algorithm for parameter estimation.

Main Methods:

  • Introduction of an input term into Cohen's existing model structure.
  • Mathematical proof of model stability using Lyapunov functions under bounded input conditions.

Related Experiment Videos

  • Development of a learning algorithm based on the stability results to minimize trajectory error.

    • Main Results:

    • The extended model successfully integrates input terms for system identification.
    • The model demonstrates stability, ensuring bounded states for bounded inputs.
    • The proposed learning algorithm guarantees convergence to optimal parameters, minimizing model-data discrepancies.

    Conclusions:

    • The enhanced model offers a novel approach for system identification with qualitative data.
    • The proven stability and convergence of the learning algorithm validate the model's practical applicability.
    • This framework facilitates a more direct and robust integration of theoretical insights and empirical data in dynamical system modeling.