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

Resonant spatiotemporal learning in large random recurrent networks.

Emmanuel Daucé1, Mathias Quoy, Bernard Doyon

  • 1Movement and Perception (UMR6559), Faculty of Sport Science, University of the Mediterranean, 163, avenue de Luminy, CP 910, 13288 Marseille cedex 9, France. dauce@esm2.imt-mrs.fr

Biological Cybernetics
|August 30, 2002
PubMed
Summary

This study introduces a two-layer neural network model that learns to predict stimuli. The system exhibits resonant behavior, enabling memory retrieval and recognition of familiar spatiotemporal signals.

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

Foveated Retinotopy Improves Classification and Localization in Convolutional Neural Networks.

Vision (Basel, Switzerland)·2026
Same author

Structure from rank: Rank-order coding as a bridge from sequence to structure.

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

L-Dopa-induced changes in aperiodic bursts dynamics relate to individual clinical improvement in Parkinson's disease.

NPJ Parkinson's disease·2025
Same author

Emergence and maintenance of modularity in neural networks with Hebbian and anti-Hebbian inhibitory STDP.

PLoS computational biology·2025
Same author

A dual foveal-peripheral visual processing model implements efficient saccade selection.

Journal of vision·2024
Same author

An exploratory computational analysis in mice brain networks of widespread epileptic seizure onset locations along with potential strategies for effective intervention and propagation control.

Frontiers in computational neuroscience·2024

Area of Science:

  • Computational Neuroscience
  • Artificial Intelligence
  • Robotics

Background:

  • Biological perceptual systems inspire novel computational models.
  • Recurrent neural networks with time delays can exhibit complex dynamics.
  • Hebbian learning rules are fundamental to synaptic plasticity.

Purpose of the Study:

  • To develop a biologically inspired neural network model for stimulus prediction and recognition.
  • To investigate the role of learning and feedback in stabilizing chaotic dynamics.
  • To explore the model's capacity for memory maintenance and real-time environmental interaction.

Main Methods:

  • A two-layer neural network with sigmoidal neurons and constant time delays was implemented.
  • A Hebbian learning rule was applied to modify network weights based on periodic stimuli.

Related Experiment Videos

  • The model's dynamics, learning effects, and resonant behavior were analyzed, including capacity assessment.

    • Main Results:

    • Learning simplified the secondary layer's chaotic dynamics, leading to stable periodic orbits.
    • A feedback connection emerged, enabling the system to predict and resonate with familiar stimuli.
    • The model demonstrated recognition of partial signals, memory maintenance, and successful real-time robotic implementation.

    Conclusions:

    • The proposed model effectively learns to predict spatiotemporal stimuli, mimicking biological perceptual systems.
    • Resonant behavior facilitates dynamic memory and recognition, highlighting the importance of feedback connections.
    • The model's high capacity and real-time capabilities suggest potential for autonomous systems.