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

Non-homogenous neural networks with chaotic recursive nodes: connectivity and multi-assemblies structures in

Emilio Del Moral Hernandez1

  • 1Polytechnic School of the University of Sao Paulo, Department of Electronic Systems Engineering, Cidade Universitaria, Av. Prof Luciano Gualberto, Sao Paulo, SP CEP 05508-900, Brazil. emilio_del_moral@ieee.org

Neural Networks : the Official Journal of the International Neural Network Society
|August 9, 2005
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same journal

Injecting reasoning into vision-language models via weight-decomposed merging.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

A Non-equilibrium thermodynamic framework for neural networks: A principled correspondence and parameter dynamics.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Dual-pathway mask ranking guided selective fine-tuning for backdoor purification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Discriminative transfer feature learning for unsupervised domain adaptation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

ARetinex-Net: Low-light image enhancement with adaptive retinex model and global illumination representation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

A Multi-Task Learning Framework with Physics Embedded for Signal Reconstruction and State Prediction in Underground Multi-Robot Systems.

Neural networks : the official journal of the International Neural Network Society·2026

This study introduces chaotic recurrent neural networks with bifurcating nodes for pattern storage. A strategy effectively minimizes pattern recovery errors, even with increased data and noise in these complex systems.

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Dynamical Systems

Background:

  • Recurrent neural networks (RNNs) are foundational for sequence processing.
  • Chaotic dynamics in neural networks offer unique computational properties.
  • Associative memory models aim to store and retrieve patterns robustly.

Purpose of the Study:

  • To investigate recurrent neural architectures utilizing bifurcating nodes with chaotic dynamics.
  • To analyze the self-organization and pattern encoding capabilities of these networks.
  • To develop strategies for enhancing associative memory performance under noisy conditions.

Main Methods:

  • Modeling neural networks with logistic recursive nodes and parametric coupling.
  • Analyzing network evolution towards spatio-temporal period-2 attractors.

Related Experiment Videos

  • Evaluating associative memory performance via pattern recovery error and basin of attraction analysis.
  • Developing and testing a strategy to mitigate performance degradation with increased stored patterns.

    • Main Results:

    • Networks self-organize into global spatio-temporal attractors encoding stored patterns.
    • Synaptic connection magnitude significantly impacts architecture performance.
    • A developed strategy successfully minimizes pattern recovery degradation as the number of stored patterns increases.
    • Mechanisms for handling asynchronous input changes and interconnecting assemblies were created.

    Conclusions:

    • Planned selection of synaptic connection scale is crucial for RPEs architectures.
    • The proposed strategy enhances the robustness of associative networks.
    • The developed architectures and mechanisms offer advanced capabilities for pattern association and processing.