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

Structurally adaptive modular networks for nonstationary environments.

V Ramamurti1, J Ghosh

  • 1Department of Electrical and Computer Engineering, The University of Texas at Austin, Austin, TX 78712-1084, USA.

IEEE Transactions on Neural Networks
|February 7, 2008
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

Isolation, characterization and evaluation of growth kinetics and multilineage differentiation of ovine ovarian mesenchymal stem cells.

Theriogenology·2026
Same author

First results from testing of the full ITER hard x-ray monitor prototype on the ADITYA Upgrade tokamak.

The Review of scientific instruments·2026
Same author

Capivasertib plus paclitaxel as first-line treatment for metastatic triple-negative breast cancer: results from the randomised, global phase III CAPItello-290 trial.

Annals of oncology : official journal of the European Society for Medical Oncology·2025
Same author

Enhancing marine magnetic anomaly interpretation with anisotropic diffusion and deep transfer learning.

Scientific reports·2025
Same author

Upgraded space and time resolved visible spectroscopic diagnostic on ADITYA-U tokamak.

The Review of scientific instruments·2024
Same author

Author Correction: Role of pinch in Argon impurity transport in ohmic discharges of Aditya-U Tokamak.

Scientific reports·2023
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 a novel neural network that dynamically adapts its structure for time-varying nonlinear systems. This adaptive architecture optimizes model selection and parameter estimation for complex environments.

Area of Science:

  • Artificial Intelligence
  • Machine Learning
  • Computational Neuroscience

Background:

  • Traditional neural networks often struggle with time-varying nonlinear systems.
  • The mixture of experts framework provides a foundation for complex modeling but requires adaptation.
  • Model selection and parameter estimation are critical challenges in dynamic system modeling.

Purpose of the Study:

  • To introduce a novel neural network architecture capable of dynamic adaptation for time-varying nonlinear input-output mapping.
  • To address the limitations of fixed architectures in modeling complex, evolving environments.
  • To improve model selection and parameter estimation through structural adaptation.

Main Methods:

  • Development of a neural network based on the mixture of experts framework.

Related Experiment Videos

  • Implementation of a localized gating network model.
  • Employing a structural adaptation procedure for growing or pruning network modules (experts) based on problem complexity.
  • Extension of batch mode learning equations to derive on-line update rules for time-varying environments.

    • Main Results:

    • The proposed network dynamically adapts its architecture to model time-varying nonlinear systems.
    • The structural adaptation procedure effectively addresses model selection, leading to improved parameter estimation.
    • On-line update rules enable the network to successfully model time-varying environments.
    • Simulation results validate the effectiveness of the proposed adaptive neural network techniques.

    Conclusions:

    • The introduced adaptive neural network offers a robust solution for modeling time-varying nonlinear systems.
    • Dynamic architectural adaptation enhances model performance and parameter estimation accuracy.
    • The on-line learning capability makes the network suitable for real-time applications in evolving environments.