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

Contextually guided unsupervised learning using local multivariate binary processors.

W A. Phillips1, Dario Floreano, Jim Kay

  • 1Centre for Cognitive and Computational Neuroscience, University of Stirling, Stirling, U.K.

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

Edible Pneumatic Battery for Sustained and Repeated Robot Actuation.

Advanced science (Weinheim, Baden-Wurttemberg, Germany)·2025
Same author

Edible aquatic robots with Marangoni propulsion.

Nature communications·2025
Same author

Instant variable stiffness in cardiovascular catheters based on fiber jamming.

Science advances·2025
Same author

Fast ground-to-air transition with avian-inspired multifunctional legs.

Nature·2024
Same author

Adaptive morphing of wing and tail for stable, resilient, and energy-efficient flight of avian-inspired drones.

npj Robotics·2024
Same author

A twist of the tail in turning maneuvers of bird-inspired drones.

Science robotics·2024
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

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

New results on prescribed-time synchronization of complex networks via intermittent control.

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

Variance-constrained multi-view ensemble broad network for imbalanced data.

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

Dynamic analysis and reliable mechanical optimization application of ring HNN effected with a memristive neuron.

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

DAFF-Net: A detection and search method for small-scale low surface brightness galaxies.

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

Quasi-synchronization for complex networks with hybrid pinning intermittent control.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

This study introduces a new methodology for multi-stream neural networks, enabling local processors to extract more coherent features across data streams. The approach enhances feature detection and learning, particularly with noisy inputs.

Area of Science:

  • Computational Neuroscience
  • Machine Learning
  • Information Theory

Background:

  • Previous work fused feature discovery and associative learning using information theory for single-stream processing.
  • Existing models extract a single coherent feature across streams using local binary processors.

Purpose of the Study:

  • To develop multi-unit local processors with multivariate binary outputs for extracting multiple coherent features across data streams.
  • To define information-theoretic objective functions and derive learning rules for these processors.

Main Methods:

  • Utilized the Ising model to define objective functions and local approximations.
  • Derived learning rules, comparing them to the BCM rule.
  • Conducted computational experiments to evaluate the methodology's properties.

Related Experiment Videos

Main Results:

  • The local methodology demonstrated required functionality for multi-stream processing.
  • Units within processors learned to specialize in different aspects of receptive fields.
  • Distributed, correlated, and damage-robust codes were produced, with competitive learning emerging in specific conditions.
  • Contextual connections facilitated extraction of cross-stream information and improved feature detection and generalization.
  • The methodology enabled learning statistical associations in distributed population codes.

Conclusions:

  • The proposed local methodology effectively extracts multiple coherent features in multi-stream neural networks.
  • Contextual guidance within the network architecture improves processing robustness and generalization.
  • This approach offers a biologically plausible mechanism for learning complex associations in neural systems.